Mental Health Services in Nursing Homes: Regulatory Oversight, Payment Policy, and Quality Improvement in Mental Health Care in Nursing Homes

In a 1986 report to Congress on improving the quality of care in nursing homes, the Institute of Medicine cited both the inappropriate use of antipsychotic drugs and physical restraints and the inadequate treatment of depression in nursing homes (3). Thus when Congress enacted the comprehensive Nursing Home Reform Act as part of the Omnibus Budget Reconciliation Act of 1987 (OBRA), it directed HCFA to take steps to ensure that unmet mental health needs among nursing home residents were addressed.

The resultant HCFA regulations require preadmission screening and annual resident review to identify patients with mental illness and ensure that they are appropriately placed in residential or treatment settings (4). The nursing homes must "ensure that a resident who displays mental or psychosocial adjustment difficulties receives appropriate services to correct the assessed problem" (5), including treatment not otherwise provided for by the state (6).

The regulations also require that periodic evaluations of nursing home residents be conducted with a standardized resident assessment instrument (7) designed to enhance the recognition of mental and behavioral symptoms that should be addressed in the treatment plan. Together, these regulations provide a clear federal mandate for the detection and treatment of mental illness in nursing homes (8).

However, there remain concerns that other federal regulations, surveyor guidelines, quality indicators, payment mechanisms, and inspections and enforcement initiatives may have the effect of undermining the regulations that are intended to ensure that mental health care is provided for nursing home residents who need it.

In response to the Institute of Medicine report, the Nursing Home Reform Act specifically directed HCFA to develop regulations to protect nursing home residents from the inappropriate use of physical restraints and from psychotropic drugs when misused as chemical restraints (5). HCFA charged the states with the responsibility for conducting periodic surveys to ascertain whether nursing facilities were in compliance with these federal regulations (9).

After these federal regulations and the nursing home survey process were implemented in the early 1990s, several studies found evidence of a substantial impact on behavioral and psychiatric treatment (10,11,12,13). Studies documenting reductions in the use of physical restraint and antipsychotic drugs and increases in prescriptions for antidepressant drugs are summarized in the article by Ryan and colleagues in this issue of the journal (14). Seigler and associates (15) examined the effect of HCFA regulations on the appropriateness of antipsychotic drug use and found that the proportion of residents receiving antipsychotic drugs who lacked a HCFA-approved indication declined from 21.3 percent to 14.6 percent. However, a retrospective study conducted at a single nursing facility revealed that 20 percent of residents whose prescription for an antipsychotic drug was discontinued or reduced in dosage subsequently had the agent resumed or its dose increased (16). This is consistent with findings from earlier discontinuation studies and suggests that a reduction in overall rates of antipsychotic drug use cannot be interpreted as an indication of across-the-board improvement in the quality of care for all residents. A more recent study of pharmacy records in eight nursing homes found that 70.9 percent of the 17.7 percent of residents receiving antipsychotic medications had a HCFA-approved diagnostic indication, 90.4 percent had documentation of appropriate target symptoms, and 90.1 percent were receiving dosages within the limits specified in the HCFA guidelines (17). These findings suggest relatively high rates of compliance with federal regulations. However, pharmacoepidemiological studies have not adequately examined the impact of compliance with the federal guidelines on symptom control, functional status, quality of life, and other outcome measures relevant to patient health and well-being.

Despite the federal mandate for assessment and treatment of patients with mental disorders in nursing homes, there is evidence that a substantial proportion of residents still have undetected psychiatric symptoms and that others do not receive the care they need (18). Borson and associates (19) examined a sample of 510 patients referred for psychiatric assessment as part of the required preadmission screening. They found that 88 percent of the sample were appropriately placed according to their care needs but that 55 percent had unmet mental health services needs and 25 percent had psychiatric disorders associated with dementia or mental retardation.

In a recent survey of administrators of nursing homes in six states, conducted by Reichman and associates (20), 47.6 percent of 899 respondents indicated that the frequency of on-site psychiatric consultation was inadequate. Thirty-eight percent of nursing home residents were judged by directors of nursing to need a psychiatric evaluation, but more than one-fourth of rural nursing homes and more than one-fifth of small nursing homes reported that no psychiatric consultant was available to them. Thus the HCFA requirement that patients receive needed mental health care did not remedy the lack of access to mental health services in nursing homes (21).

In addition to the federal requirements for detection and treatment of mental disorders in nursing homes, changes in Medicare Part B payment rules initially encouraged the provision of mental health services. During the late 1980s, the limits on payments for outpatient psychiatric services were raised and ultimately eliminated. Also, in 1990, the Medicare Part B psychiatric benefit was expanded to allow licensed clinical psychologists and certified social workers to bill Medicare for mental health services. These factors led to a substantial increase in Medicare payments for mental health services in nursing homes during the early and mid-1990s. However, the extent to which the increased spending improved access or addressed unmet needs for mental health care is not known.

Subsequent developments threatened to undermine efforts to improve the quality of mental health care in nursing homes (22). First, the federal Balanced Budget Act of 1997 repealed federal standards for reimbursing nursing home care under the Medicaid program, which covers 68 percent of nursing home residents and more than 59 percent of nursing home costs. This measure gave states freedom to set payment rates, resulting in significant disparities across states. In 1998, average Medicaid nursing home reimbursement rates fell as low as $62.58 a day in Nebraska. Second, the 1997 law altered reimbursement methods for nursing homes, yielding large budget savings for Medicare but resulting in the withdrawal of substantial resources from long-term care.

In addition, beginning in 1999, social work services were required to be furnished as one of many services bundled together under payments made directly to nursing facilities. Certified social workers are no longer allowed to bill Medicare for psychiatric services delivered in skilled nursing facilities. This means that payment for mental health care provided by social workers comes out of the nursing home's per diem reimbursement. Considered together, these are disincentives for providing nursing home-based services required for the "psychosocial well-being" of nursing home residents. This is an example of contradiction across policy domains: regulatory policy requires mental health assessment and treatment, whereas payment policy undermines this goal by allowing states to underfund nursing homes.

Although the states have a role in monitoring compliance with the federal regulations through the nursing home survey process, the Office of Inspector General is also charged with overseeing the provision of mental health services to nursing home residents. The statutory mission of the office is to protect the integrity of Department of Health and Human Services programs, including Medicare and Medicaid.

During the past six years, the Office of Inspector General has conducted two studies of mental health services provided in nursing homes. The first, released in May 1996 (1), was prompted by a tripling of Medicare payments for nursing home mental health services. The increased payments coincided with an increase in the number of nursing home residents identified as having mental disorders, which occurred after the introduction of the preadmission screening and annual resident review and the "minimum data set," which is a standardized instrument that includes evaluations of mood, cognition, behavior, and functional status (5). They also coincided with the expansion of Medicare Part B coverage of mental health services to include direct billing by clinical psychologists and certified social workers. In response to allegations that Medicare was being billed for unnecessary or inappropriate services, the Office of Inspector General conducted a medical record review of 397 beneficiaries who received care billed under five specific codes for psychiatric diagnostic interviews, psychological testing, and individual or group psychotherapy. Thirty-two percent of these services were deemed medically unnecessary—that is, the patient's condition did not warrant the treatment or the patient was unable to benefit from treatment—and 15 percent were "questionable."

The second study, released in January 2001, focused on the appropriateness of Medicare Part B payments for psychiatric services (2). This study was based on a medical record review of 365 services billed under any one of five codes for psychological testing, group psychotherapy, and individual psychotherapy with and without medical evaluation and management. Twenty-seven percent of services billed under these five codes were deemed to be medically unnecessary, in most cases on the basis of a determination that the patient's cognitive capacity precluded his or her benefiting from the treatment or that the frequency or duration of the service was excessive. In addition, the study found that 9 percent of services lacked any psychiatric documentation and 3 percent were questionable, primarily because of incomplete documentation or cognitive deficits that would likely preclude treatment benefit. As with the first study, psychological testing, provided mostly by clinical psychologists, was the service most often found to be medically unnecessary, primarily because of lack of need or excessive length or frequency of testing. Least problematic was individual psychotherapy with medical evaluation and management (code 90817).

It must be kept in mind, however, that these reports contained information only about the specific mental health service codes studied. The studies focused on the psychological services and the psychological testing services that accounted for a majority of the nursing home mental health care reimbursed by Medicare or that were thought to be most prone to abuse. Neither of the studies examined services billed under other psychiatric service codes specifically for nursing home care (for example, initial evaluation code 99303 or follow-up visit codes 99311, −312, −313). Thus inferences cannot be drawn from these studies about the appropriateness of the entire range of psychiatric care currently provided in nursing homes. The studies were not designed to determine the effectiveness or efficiency of services delivered, the extent to which nursing home residents benefited from services not previously available to them, or how many residents were still not receiving needed care. Although it is possible that some proportion of increased Medicare spending might reflect improvements in access to mental health services or in quality of care for some nursing home residents, this has not yet been evaluated by health services or outcomes research.

Research on reimbursement and its potential impact on quality of care has been focused on the level of expenditures for health services and the method of payment. A few studies have shown that costs or expenditures are positively related to staffing intensity and that professional staffing has a positive and significant relationship to outcomes-based quality measures (23,24). However, it has not been demonstrated that higher reimbursement levels have a significant direct impact on quality as measured by outcomes of care. Neither has the relationship between the method of payment—flat rate, prospective payment, use or type of case mix adjustment—and access to psychiatric services or quality of mental health care been studied. Although there does not seem to be a simple relationship between reimbursement and access and quality, it is logical to expect that there is some minimal level of reimbursement below which it is either difficult or impossible for nursing homes to provide an adequate level of care. The most recent Institute of Medicine report calls for research supported by the Department of Health and Human Services to address the effects of payment policy on access to services and quality of care (20).

The Surgeon General's 1999 report on mental health recognizes the need for an integrated approach to the provision of mental health care for older adults. It asserts that a bundled payment system, based on comprehensive assessments, would provide the needed resources and would challenge providers to do what is necessary to provide the care (25). For such a system to work effectively in the nursing home setting, the payer must provide payments sufficient to enable the facility to provide an acceptable level of care, and the facility must assess patients and provide needed treatment. Accountability is thus required of both the payer and the provider.

One of the essential elements of such a payment system, the use of a resident assessment instrument, has already been mandated by the federal regulations and implemented by HCFA. For skilled nursing facilities, the assessment process is a requirement for participation in Medicare and serves as the basis for both internal and external quality assurance. All patients in nursing homes must undergo periodic comprehensive assessments using the minimum data set. Responses on the minimum data set that indicate deficits or changes in health status trigger second-stage assessments with resident assessment protocols that assist staff members in determining the need for changes in treatment.

The data acquired through the minimum data set are also used by HCFA to guide Medicare payments for nursing home care up to the first 100 days. Data derived from the assessment of patients' needs drive the level of reimbursement to cover the services required to meet those needs. The prospective payment system for skilled nursing facilities uses the resource utilization groups that include categories for patients with mental illness. Information from the minimum data set related to mental disorders is used to assign individual patients to the proper resource utilization group for the purposes of payment. Because this payment system is based on Medicare's historical costs for 1995, there continue to be questions about how much additional care and enhancement of quality may be needed currently and how much it should cost. Nevertheless, resource utilization groups are an example of a method for measuring patients' needs and dispensing payments. The expectation is that quality improvement activities over time will help address the issues of cost and quality.

Ultimately, a payment system can be used to drive quality improvement (20,23). Assessment, survey, and payment data can be used to analyze the quality of care and, in turn, feedback can inform refinements of the payment system. In 1999 HCFA introduced quality indicators, derived from the minimum data set, to enable facilities and surveyors to compare individual facilities within the same state and to use the results to identify and address potential quality problems (26). Currently 24 quality indicators within 11 domains are used, including behavioral-emotional problems, cognitive patterns, and psychotropic drug use. Whenever a review in any of these areas results in a citation of deficiency, a plan of correction must be developed and submitted for approval. This system is a first step in monitoring quality of care, although the face validity of some of the quality indicators has been questioned and the results of quality surveys may be difficult to interpret (27). Nevertheless, this is a potent mechanism by which the assessment data required by the payment system can drive quality improvement efforts.

During the past 15 years, federal regulations, survey and inspection programs, and payment policies have presented shifting—and often conflicting—incentives and disincentives for the provision of mental health services in nursing homes in the United States. During that period, some reductions in abusive and inappropriate treatment of nursing home residents were achieved, which suggests that payment policy and regulatory oversight can be used effectively to promote quality improvement. However, substantial improvements in access and quality of mental health services in nursing homes are still needed. Specific recommendations contained in recent reports issued by the Surgeon General and the Institute of Medicine highlight the need for refinements in the assessment process, the use of outcomes-based quality measures, and payment policies and reimbursement mechanisms that provide incentives for improved access and quality. All these elements—regulatory oversight, payment policy, and quality improvement—must be coordinated to promote humane treatment in nursing homes, including access to appropriate psychiatric care.

This research was supported in part by grant MH-52129 from the National Institute of Mental Health and the VA VISN 4 Mental Illness Research Education Clinical Center.

Dr. Streim is with the section on geriatric psychiatry at the University of Pennsylvania, 3535 Market Street, Philadelphia, Pennsylvania 19104 (e-mail, [email protected]), and with the Mental Illness Research, Education, and Clinical Center at the Philadelphia VA Medical Center. Ms. Beckwith is a health care policy consultant in Washington, D.C. Ms. Arapakos, Ms. Banta, and Ms. Dunn are with the Office of Inspector General of the Department of Health and Human Services in New York City. Mr. Hoyer is with Centers for Medicare and Medicaid of the Department of Health and Human Services in Washington, D.C. This article is part of a special section on mental health services in long-term-care facilities.