An Agency-Based Representative Payee Program and Improved Community Tenure of Persons With Mental Illness

Severely mentally ill individuals often have difficulties in managing money from their Social Security payments. Lack of financial skills can cause problems in paying rent and purchasing other necessities, which can lead to homelessness. The stress of homelessness may then exacerbate symptoms, resulting in rehospitalization. Representative payee programs help individuals with serious mental illness budget Social Security payments for necessities.

Representative payee programs provide money management services, which have several important roles. They protect the client by preventing harm due to loss or misuse of funds needed for survival. They facilitate maintenance of the client in the community by stabilizing the client's income flow and housing and by providing skills training in money management, such as budgeting, as well as advocacy to secure financial entitlements. Services include paying for rent, utilities, medical expenses, and food. The representative payee may be a staff member from a community agency, a family member, or a friend.

Very little evaluative research has been conducted on representative payee programs for mentally ill individuals (1), and even fewer studies have examined use of an agency as the representative payee (2). Stoner (2) reported a descriptive study in which agency staff functioned as representative payees for mentally ill clients. Money management services were provided by Skid Row Mental Health Service in Los Angeles at a drop-in day shelter for 89 homeless mentally ill persons. Baseline information, as well as follow-up information at one year after enrollment, was collected using a monthly client record form completed by case managers. Findings were fairly positive on several measures. After one year in the program, 77 percent of the clients reported no days of homelessness, 82 percent were living within their monthly allocation, 9 percent had been crime victims, and 9 percent had been arrested. At baseline, 50 percent of the group had reported that they either were crime victims or had been arrested in the previous year.

Staff ratings of clients' cooperation with treatment indicated an improvement; ratings of "very cooperative" increased from 45 percent of the clients at enrollment to 81 percent after the first year (2). Data on hospitalizations were not reported. Obviously, a control group would have strengthened the assumption that the results were due to the intervention. Nevertheless, the study suggested a link between money management services and an increase in clients' ability to live within their means. The author concluded that money management is a critical element of care (2).

Further research is needed on this important service. In addition, no empirically based guidelines are available for referring mentally ill individuals to representative payee programs. The purpose of the study reported here was to determine what criteria are used in referring clients to representative payee programs and to examine what effect participation in the program has on hospitalization.

Methods

Setting and sample

A representative payee program was developed in 1981 by Community Counseling Centers of Chicago, an urban mental health center on Chicago's North Side. Representative payee services are provided within the context of other intensive services, such as case management and day treatment.

The sample consisted of 56 persons enrolled in the representative payee program. Most had schizophrenia (21 clients, or 38 percent) or schizoaffective disorder (16 clients, or 29 percent). The others had bipolar disorder (ten clients, or 18 percent) or other diagnoses (nine clients, or 16 percent), including depression, impulse control disorders, and personality disorders.

The majority were male (38 clients, or 68 percent), and relatively few were married (six clients, or 11 percent). The mean±SD age of the sample was 42±10 years. Two-thirds were Caucasian (37 clients), and the remainder were African American (ten clients, or 18 percent), Hispanic (six clients, or 11 percent), Asian-Pacific Islander (one client), or other (two clients).

The majority of representative payee clients received Supplemental Security Income (SSI) (36 clients, or 64 percent). The remainder received Social Security Disability Insurance (20 clients, or 36 percent). Forty-five clients (80 percent) were mandated to representative payee status, usually by their physicians because they had demonstrated financial incompetence; for example, some had had periods of homelessness because of failure to pay rent.

Design and intervention

The 56 severely mentally ill individuals in this retrospective pre- and postintervention study were enrolled in the representative payee program for at least one year and had used services at the agency for at least one year before enrolling in the program. The enrollment period ranged from 1985 to 1995.

Social Security payments are managed by a team consisting of case managers and banking staff who function as representative payees. A section of the Community Counseling Centers of Chicago facility has been set aside as a bank, complete with tellers and a security guard. Case managers and clients together develop a budget that includes weekly or daily cash disbursements that are handled by the banking staff. Rent and utility bills are usually paid by the banking staff as well.

Assistance with locating housing is also an important part of the representative payee program. Landlords who are familiar with the agency are willing to provide housing to clients with poor credit records when the agency is acting as representative payee. Although budgets are individualized, a common pattern is for the initial budget to include a daily or biweekly allowance, with a gradual shift to a weekly or monthly allowance as clients improve their money management skills. Skills training is aimed at increasing clients' independence. Clients may, for example, eventually take charge of bringing the rent check to the landlord rather than having the agency's bank pay the landlord directly.

Measures

One purpose of the study was to determine specific criteria that were being used in referring clients to the representative payee program. A literature review, focus groups, case studies, and an expert panel identified several criteria on the basis of which clients might be referred to a representative payee program. For example, clients referred to such programs often have severe mental illness with concurrent substance abuse or dependence, frequent or long-term hospitalizations, a history of homelessness, a lack of financial skills, a history of impulsive spending, and no money for basic necessities such as food, rent, and clothing. Clients who need representative payees typically do not accept the fact that they need treatment and are a danger in their residence. They have a pattern of hospitalization following receipt of a Social Security check and a history of victimization by others.

A chart review was conducted to determine the extent to which these criteria were present among clients referred to the program. Interrater agreement for the chart review was .87. The major focus of the chart review was the client's individual treatment plan, which includes problems that are considered significant. Judgments about the presence or absence of the criteria were based on chart notes; for instance, a note might state, "Client lacks money management skills," or "Client has been abusing alcohol and cocaine." Frequent hospitalization was defined as having three or more hospitalizations in one year. Long-term hospitalization was defined as a stay of a year or longer.

Impulsive spending refers to clients' spending a large proportion of their money on items that are not basic necessities, leaving too little to get by on for the rest of the month. Victimization by others refers to situations in which the client is pressured to spend money inappropriately. For example, drug dealers know when Social Security checks are due and sometimes push clients to spend their money on drugs, or so-called friends may ask the client for a large portion of the check. Lack of acceptance of the need for treatment refers to clients who do not follow through with key treatment activities, such as keeping clinic appointments and taking medication.

The results of an analysis of other differences between clients who were and were not referred to the representative payee program are reported elsewhere in this issue (3).

Hospital use was calculated for the year before referral to the representative payee program and for the year of involvement in the program. This information was obtained from the Illinois Department of Human Services, which maintains a statewide database with records of hospital stays for all state hospitals. Information about private hospital stays funded by Medicaid was available only for the 33 clients enrolled in the representative payee program from fiscal years 1992 to 1996, because the Medicaid database was not available until 1992.

Analysis

Frequency distributions were used to describe the proportion of the various criteria used to refer clients to the representative payee program. Repeated-measures t tests were used to examine pre- and postenrollment changes in number of days hospitalized, the number of state hospitalizations for the entire sample, and the number of Medicaid-funded private hospitalizations for the 33 clients for whom data were available.

A Bonferroni adjustment for the use of four significance tests required a probability level of .01 for statistical significance.

Results

Enrollment criteria

The most common condition associated with enrollment in the representative payee program was comorbid substance abuse or dependence (27 clients, or 48 percent). Extensive institutional care was also a factor, with many clients having frequent hospitalizations (18 clients, or 32 percent), long-term hospitalizations (12 clients, or 21 percent), or a pattern of hospitalization following receipt of their SSI check (three clients, or 5 percent). A representative payee was also seen as being important for clients with a history of homelessness (17 clients, or 30 percent).

Another important factor in referral was a lack of financial skills (16 clients, or 29 percent); it sometimes resulted in impulsive spending (nine clients, or 16 percent) and inadequate money for basic needs such as rent (six clients, or 11 percent) or food (six clients). Other relatively frequent reasons for referral included the client's being a danger in his or her residence (12 clients, or 22 percent) and the need to motivate the client to participate in treatment (ten clients, or 18 percent).

Reduced hospital stays

The number of days in state hospitals was dramatically reduced during the year of program participation, from a mean±SD of 67.9±104.8 days to 6.8±17 days. This difference was highly significant (repeated-measures t test= 4.36, df=55, p<.001). The number of state hospitalizations also decreased significantly from 1.1±1.7 to .3±.7 (repeated-measures t test=3.94, df=55, p<.001).

A similar pattern of reduced hospital stays occurred within the subgroup of 33 clients for whom data were available on both state hospitalizations and Medicaid-funded private hospitalizations. The number of days in private and state hospitals decreased from 96.7±115.2 days to 14.5±26.7 days (repeated-measures t test=4.01, df=32, p<.001). The number of combined state and private hospitalizations for this subgroup also decreased, from 1.7±2 to .9±1.7 (repeated-measures t test=2.79, df=32, p=.009).

For number of days hospitalized, the only differential effect among clients in the representative payee program involved age. Younger clients were more likely to spend fewer days in the hospital (R=-.39, p<.01). No differences in number of days hospitalized were associated with other demographic variables or the presence or absence of substance abuse.

Discussion and conclusions

The most important finding was the more than ninefold reduction in the number of days clients spent in the state hospital—from a mean of 68 days to seven days—in the year after enrollment in the representative payee program, with a parallel reduction in the number of days in state and private (Medicaid-funded) hospitals, from 97 days to 15 days. The number of hospitalizations also decreased. These findings suggest that the representative payee program was quite successful in maintaining clients in the community.

Limitations

In the absence of a control group, it is important to recognize that decreases in hospitalization over time could be due to factors other than participation in the representative payee program. The most important alternative explanation would be the possibility of a trend toward less hospital use over time because of continuing deinstitutionalization. State-level data were examined to assess this possibility. State hospital data from 1985 to 1995 do reveal a decrease of 44 percent in hospital admissions. In comparison, clients in the representative payee program decreased their mean number of hospital admissions by 73 percent—from 1.1 to .3.

State trends for combined public hospital and Medicaid-funded psychiatric hospitalizations also provide an informative contrast because of shifts in utilization from state hospitals to Medicaid-funded providers. When both sources were included, the rate of psychiatric admissions in Illinois actually increased by about 16 percent during a comparable time period (1986 to 1994), while the number of hospital days almost tripled. By contrast, for clients in the representative payee program for whom state and Medicaid data were available, both the mean number of hospital admissions and the mean number of days hospitalized decreased significantly during the year of program participation—from 1.7 to .9 admissions and from 97 to 15 days.

Although a controlled study would provide more conclusive evidence, these data suggest that the changes observed in this study were not due to secular trends. Our controlled study, which is in progress, will clarify this issue further. It will involve a prospective comparison of clients with representative payees and a control group.

Development of criteria

The study reported here also provided an important first step in developing criteria for referring clients to representative payee programs. The reliance on retrospective chart reviews to identify referral criteria is a limitation of the study, as case managers may have had reasons for making referrals that were not noted in the chart. However, given the complete absence of any systematic study of this issue, these preliminary findings should provide some guidance to practitioners about which clients may be in need of a representative payee program.

The most common reason for referral to the program was comorbid substance abuse, which characterized nearly half of the group. The fact that in terms of hospital stays, clients with substance abuse had outcomes similar to those of the other clients may support the value of the intervention for this population.

Program efficacy

The nature of this study did not permit a component analysis to determine what part of the representative payee program was helpful. The helpful element is unlikely to be social support from counselors, because the clients were already engaged in treatment at the facility for a year before enrollment in the program. The core function of the program is the disbursement of funds based on a carefully defined budget that ensures that housing is provided and that basic survival needs are met.

Both this study and related research suggest that this approach is essential for maintenance in the community. The emphasis on meeting survival needs and on finding and keeping housing is also central to assertive community treatment. Models based on assertive community treatment have demonstrated reduced use of psychiatric hospitals (4,5). Positive outcomes have also included improvements in finances due to newly obtained entitlements, better resource utilization, and improvement in meeting survival needs (6,7).

Autonomy and beneficence

Because the representative payee program limits a client's financial autonomy, ethical issues arise. Beneficence and autonomy are two core principles underlying ethical judgments (8). As the term suggests, beneficence refers to the need to make decisions that are in the individual's best interests. Autonomy concerns the need to respect a person's right to self-determination and to empower a client to make the decisions that greatly affect his or her life.

When individuals are severely disabled, they may have impairments that interfere with their decision-making and self-care skills in ways that put them at risk for harm. In such situations, the ethical obligation to protect a client's welfare can conflict with the client's right to self-determination (9). In paternalistic beneficence, the clinician's view of benefits and harms is different from that of the client, and the clinician's judgment prevails (8).

Opinions abound about the relative merits of autonomy versus beneficence on any given issue, but research on stakeholders' views is sparse indeed. In one European study, opinions about compulsory commitment were sought from both committed and voluntary inpatients as well as their relatives, health and mental health staff, and a sample of the general public (10). Support for medical paternalism was strong, and most viewed doctors, not legal authorities, as the best people to decide about commitment. It is not known to what extent these findings would generalize to the United States, as no such American studies were identified. In the absence of a social consensus on this and most other ethical issues, it is necessary for the time being to rely on the opinions of individual stakeholders together with their interpretations of ethical principles.

Paternalistic beneficence ranges from extensive to limited (11). It is extensive when it is applied to all of the major aspects of a client's life; it is selective when interventions concern protecting the client in specific areas, with substantial limitations of liberty; and it is limited when the client is protected from harm without having his or her basic liberties interfered with. Advocates of the need for long-term asylum can be said to favor extensive paternalistic beneficence for severely mentally ill individuals, who are seen as being incapable of functioning in the community (12). An example of selective beneficence would be the use of involuntary commitment for a limited time to prevent clients from suicide.

Money management services, such as representative payeeship, have been classified as limited paternalistic beneficence (12). Through a representative payee, the client is protected from harm due to loss or misuse of the resources necessary for basic survival needs such as food and shelter (13). From both a legal and an ethical perspective, the provision of representative payee services must be justified by a clinical judgment that the client is incapable of managing funds (14). Based on a series of case studies of representative payeeship, one study concluded that the benefits of preventing deterioration and hospitalization by ensuring that basic survival needs are met provide a strong ethical justification for limiting a client's autonomy (13).

The representative payee program of the Community Counseling Centers of Chicago attempts to enhance client autonomy as much as possible. Although participation in the program is mandated for the majority of the agency's clients, case studies and focus groups suggest that clients view help with securing housing and paying for rent and other necessities as beneficial. The representative payee program provides skills training to increase clients' autonomy in an attempt to balance the need to protect clients from the harmful effects of mismanaging money. Budgets are individualized, and they change over time to reflect improvements in clients' money management skills.

If clients are dissatisfied with the agency's performance as the representative payee, they may petition the Social Security Administration to change their payee to a family member, friend, or another agency. However, anecdotal evidence from focus groups suggests that most clients are highly satisfied with the program and see it as beneficial. Although the ultimate goal is for clients to become independent of the program, relatively few attain this level of recovery.

Autonomy issues also arise when contingencies are used. For example, clients are expected to see their case managers before receiving their allowance at the bank. This requirement may be viewed as an ethical issue. The intent is, of course, to enhance the likelihood that the treatment plan will be effective, not to deny the client money for basic needs. Stoner's Skid Row Mental Health Service study (2) showed a 36 percent increase in cooperation with treatment. Clearly, a balance between paternalism and autonomy needs to be struck.

Future research

This report presents preliminary findings from our program of research on money management services. Further research is greatly needed. Besides decreased hospitalization, other outcomes of representative payee programs need to be examined, such as decreases in substance abuse and homelessness. A history of homelessness was another important reason for referral to the program, which played a role in 33 percent of cases. Persons with severe mental illnesses, such as schizophrenia, have a 25 to 50 percent risk of homelessness, which is ten to 20 times the risk of homelessness for the general population (15). Our controlled study in progress will further examine substance abuse and housing stability.

Acknowledgements

This study received support from a University of Chicago Home Health Care research grant (Daniel J. Luchins, M.D., principal investigator) and from grant SM51945 from the Substance Abuse and Mental Health Services Administration (Kendon J. Conrad, Ph.D., principal investigator).

Dr. Luchins, Dr. Hanrahan, and Ms. Savage are affiliated with the department of psychiatry at the University of Chicago, 5841 South Maryland Avenue, MC3077, Chicago, Illinois 60637 (e-mail, [email protected]). Dr. Luchins is also affiliated with the Illinois Department of Human Services. Dr. Conrad and Dr. Matters are with the School of Public Health of the University of Illinois at Chicago. Dr. Shinderman is affiliated with Community Counseling Centers of Chicago. This paper was presented at the annual meeting of the American Psychiatric Association held May 17-22, 1997, in San Diego.