Screening for Mental Illness in Nursing Facility Applicants:
Understanding Federal Requirements

IX. Current Issues and Concerns

Because data specific to PASRR are limited, it is difficult to assess the nature of PASRR's impact on identifying individuals in need of mental health services. Organizations such as the American Psychiatric Association and the American Association for Geriatric Psychiatry expressed support for the program in their 1998 joint testimony to the Committee on Improving Quality in Long-Term Care. The same organizations, however, believe the program has made limited progress in providing the target population with necessary care (Colenda et al., 1999). Current concerns include the following:

Lack of Funding for PASRR Screens, Specialized Services or Alternative Treatment

Expenditures for PASRR determinations (an administrative expense) are eligible for 75 percent Federal financial participation. Level II determinations--the PASRR determination of whether an individual with SMI requires (1) nursing facility services and (2) specialized services--must be conducted by State Mental Health Authorities. However, the State share may come from any State agency. The only prohibition is that it may not come from another Federal funding source.

Funding PASRR screens and specialized treatment is a challenge for States. If individuals need specialized services to treat their mental illnesses, the State Medicaid Agency ultimately is responsible for providing or arranging for the provision of those services. Federal reimbursement is available for specialized services that are covered under the Medicaid State plan. Although the Federal Government will match State expenditures on meeting PASRR requirements, PASRR only evaluates individuals to determine the need for nursing facility services and specialized services. Finding and funding alternative placement is not a PASRR function.

Statutory Exemptions for Dementia

When OBRA 1987 was initially proposed, professional organizations such as the National Association of State Mental Health Program Directors were concerned that PASRR would force removal of residents with dementia from nursing facilities. To ensure that such individuals were not denied nursing facility care, they recommended that PASRR apply only to individuals with serious mental illness (Emerson Lombardo, et al., 1996). As a result, Congress amended the law, constricting the definition of mental illness to SMI and does not include persons with dementia unless the person has a primary diagnosis of SMI. Because States are not required to conduct Level II screens on these populations, the statutory definition of mental illness inadvertently has failed to identify the need for specialized services for many individuals who have Alzheimer's disease or related disorders. Once individuals are diagnosed with dementia from a Level I identification screen, they are not required to go through a Level II PASRR screen unless it is suspected the person has a primary diagnosis of a serious mental illness. While nursing facilities are required to provide mental health services of lesser intensity than specialized services to residents who need them, the absence of a Level II evaluation reduces the possibility that they will receive any intensive treatment needed to address symptoms of the disease. Yet several studies have documented that individuals with dementia can benefit from mental health services (Bazelon Center, 1996; Emerson Lombardo, 1994; Emerson Lombardo et al., 1996; Streim, 1995).

Lack of Clear Definitions in the Legislation

The ambiguous language contained in the legislation regarding serious mental illness and specialized services has been a concern since the enactment of PASRR (Bazelon Center, 1996; Robinson, 1990). Although Federal minimum requirements for diagnosing serious mental illness exist, States are allowed to administer their own instruments to screen for suspected mental illness--instruments that might establish broader guidelines than Federal statute. Hence, an individual might be defined as having a serious mental illness in one State but not in another. The broad definition of "specialized services" leads to a similar situation--there is wide variation in the types of services that States provide to people with serious mental illness.

Barriers to Meeting PASRR Objectives

With few resources to monitor States and limited statutory penalties except closing a facility, denying payment, or imposing fines, CMS has had difficulty enforcing PASRR. Consequently, advocates and policymakers suspect that PASRR is not having the full intended results (Bazelon Center, 1996; Borson et al., 1997; Marek et al., 1996; Sherrell et al., 1998).

Because PASRR is a Medicaid program that is designed by each State within broad Federal guidelines, little consistency among States exists. Furthermore, because PASRR is unique in that it requires coordination between State agencies that are not accustomed to working together, there are communication problems and lack of consistency within States as well. Such inconsistency can pose problems, affecting everything from a State's ability to measure outcomes to a hospital's understanding of the process when it serves individuals from more than one State (Sherrell et al., 1998; SSWLHC, 1995). For example, Sherrell and colleagues (1998) found that Illinois relied on 27 different PASRR agencies, each contracting with its own independent consulting firm to conduct Level II screens. Inconsistencies among nursing facilities in Chicago, a product of disparate measurement instruments and the structures of final reports, hindered the researchers' ability to draw meaningful conclusions about PASRR's outcomes. In addition, Sherrell and colleagues assert that psychologists conducting Level II screens were unfamiliar with the resources of the nursing facilities. Therefore, many of their recommendations for treatment were either inappropriate or not available in nursing facilities. For example, although the primary symptom identified in the Level II screens was social withdrawal, 64 percent of individuals screened were placed on psychotropic medications that were not helpful in managing withdrawal. Moreover, treatment recommendations did not vary by diagnosis, age, dementia, level of independence in daily activities, or symptoms (Sherrell et al., 1998).

Inefficient Utilization of the Expertise of Mental Health Professionals

The question of who provides mental health services to nursing home residents remains controversial in the current debate over PASRR. Emerson Lombardo and colleagues (1996) found that psychiatrists and other mental health professionals must dedicate their time to evaluation and medication management, leaving them unable to provide therapy and other more direct modes of treatment (Emerson Lombardo et al., 1996). In an unpublished study, Shea et al. (1995) found that only 29 percent of nursing home residents were receiving mental health services from a mental health professional (Emerson Lombardo et al., 1996). Some researchers say that the underlying problem limiting the use of psychiatrists and psychologists is low reimbursement rates under Medicare and Medicaid. Although there have been changes in Medicare reimbursement to encourage utilization of psychologists, Shea et al. (1995) found that fewer than 5 percent of residents in treatment were receiving services from psychologists (Emerson Lombardo et al., 1996). Furthermore, there is concern over the lack of knowledge about mental health issues among nursing facility staff. Unless staff members are trained in mental health issues, they will be unable to provide an environment conducive to improving each resident's mental health (Emerson Lombardo et al., 1996).

Administrative Burden and Cost

A study by Marek and colleagues (1996) reports that government officials and nursing facility staff are concerned by increased paperwork, administrative burden, and costs associated with OBRA 1987. Increased administrative costs are seen as effectively reducing the resources available for direct care. A study by SSWLHC (1995) indicates that the PASRR process is redundant of other nursing facility requirements to conduct assessments (e.g., RAI/MDS) of applicants after admission. Like PASRR, RAI/MDS collects information on psychiatric diagnosis. SSWLHC argues that this process alone is adequate in ensuring appropriate nursing facility placement, however, this view disregards the reality that the nursing facility resident assessment, under the minimum data (RAI/MDS) occurs after the admission when resident protections are in effect that provide the resident a 30-day notice and limit the reasons for which a facility can discharge or transfer a resident involuntarily. Regulations also provide the resident with the right to an appeal, which would extend an inappropriate nursing facility placement.

Prompted by concerns about high administrative burden and information redundancy, some organizations argue that PASRR should be repealed. However, it is not clear that the data collected in RAI/MDS, albeit completed after admission by nursing facility staff that often has no MH training, would be sufficient for determining the appropriate placement needs for mentally ill individuals. In their joint testimony to the Committee on Improving Quality in Long-Term Care, the American Psychiatric Association and the American Association for Geriatric Psychiatry note that RAI screens may not be useful in indicating need for specialized services or developing quality monitors (Colenda et al., 1999). More research is needed to determine the extent to which other assessment instruments would be able to replace PASRR.

With regard to cost, SSWLHC (1995) reported that no cost/benefit analysis of PASRR has been performed. Studies that measure costs related to PASRR for particular areas document variation in the level of spending between States. For example, States responding to the Bazelon Center survey demonstrated an average of $2.10 million per State for individuals with SMI in 1991 and $2.13 million per State in 1993. Yet Texas exhibited spending levels of $12 million and $26 million alone for those 2 years, respectively (Bazelon Center, 1996). According to the SSWLHC (1995) survey of State Mental Health Authorities, program directors reported an average expenditure of $373 per Level II screen in 1989 and $336 per Level II screen in 1991. Neither study examined cost per individual in need of specialized services, although such information would be useful in determining whether the costs outweigh the benefits of the legislation.

Delays in Placement

Hospitals and NF staff report concerns over placement delays caused by the PASRR screening process (SSWLHC, 1995). According to PASRR regulations, individuals cannot be placed in nursing facilities until after the screens have been completed and must be made in writing within an annual average of 7 to 9 working days of referral. Placements can be initiated earlier by telephone or electronic authorization. However, the SSWLHC study documents that an individual requiring both Level I and Level II screens can be delayed for as long as 3 weeks for a final determination. The SSWLHC survey found that Level II screens delayed hospital discharge by between 0 and 24.5 days, with an average of 8.3 days. The high level of variation in these findings indicates that hospital discharge delays differ from one State to another and between one hospital and another. Because each State designs its own PASRR program, some are more or less efficient than others. Also, hospitals that begin discharge planning at the time of admission; conduct a Level I at admission if it appears nursing facility placement is likely; and schedule a Level II early, have significantly fewer discharge delays than hospitals that wait until the time of discharge to perform a Level I. Advocates of PASRR reform, however, argue that placement delay problems are primarily due to the unnecessary complexity of the PASRR process.

Table of Contents | Previous | Next

Screening for Mental Illness in Nursing Facility Applicants: Understanding Federal Requirements