The American Psychiatric Association (APA) has updated its Privacy Policy and Terms of Use, including with new information specifically addressed to individuals in the European Economic Area. As described in the Privacy Policy and Terms of Use, this website utilizes cookies, including for the purpose of offering an optimal online experience and services tailored to your preferences.

Please read the entire Privacy Policy and Terms of Use. By closing this message, browsing this website, continuing the navigation, or otherwise continuing to use the APA's websites, you confirm that you understand and accept the terms of the Privacy Policy and Terms of Use, including the utilization of cookies.

×

Abstract

Objective:

The liberty of individuals who receive Social Security disability payments is constrained if they are judged incapable of managing their payments and are assigned a payee or conservator to manage benefit payments on their behalf. Conversely, beneficiaries’ well-being may be compromised if they misspend money that they need to survive. Several studies have shown that determinations of financial capability are made inconsistently and that capability guidelines appear to be applied inconsistently. This article describes ambiguities that remained for individuals even after a comprehensive assessment of financial capability was conducted by independent assessors.

Methods:

Trained, experienced assessors rated the financial capability of 118 individuals in intensive outpatient or inpatient psychiatric facilities who received Social Security Disability Insurance or Supplemental Security Income.

Results:

Ten individuals’ cases were determined to be difficult to judge. Six sources of ambiguity were identified by case review: distinguishing incapability from the challenges of navigating poverty, the amount of nonessential spending that indicates incapability, the amount of spending on harmful things that indicates incapability, how to consider intermittent periods of capability and incapability, the relative weighting of past behavior and future plans to change, and discrepancies between different sources of information.

Conclusions:

The cases raise fundamental questions about how to define and identify financial incapability, but they also illustrate how detailed consideration of beneficiaries’ living situations and decision making can inform the difficult dichotomous decision about capability.

In 2011, 3.7 million people with psychiatric disabilities who were judged unable to work received monetary benefits from the Social Security Disability Insurance (SSDI) and Supplemental Security Income (SSI) programs (1). When used as intended, Social Security benefits help provide disabled individuals with money for food, housing, or clothing (herein referred to as basic needs) that they might not be able to afford. However, incidents of benefits misspending, which have been described in the literature, including use of disability benefits to purchase alcohol or drugs and excessive spending during acute psychotic, manic, or depressive episodes, have caused beneficiaries to depend on others for basic needs or suffer loss of their benefits (2,3). Such misspending is particularly common among individuals with mental illnesses that impair their cognitive abilities, judgment, and the ability to resist financial exploitation (37). Independent financial management may be further compromised when individuals with mental illness have concurrent substance use disorders (5,6,8).

Literature addressing capability among individuals with mental illness often focuses on their capacity to provide informed consent for treatment (9) or research participation (10). There is limited literature addressing financial capability of people with mental illness (3). Clinicians, courts, Social Security Administration (SSA) claims officials, and others involved in determining whether beneficiaries are incapable of managing their finances provide guidelines for such determinations, but these guidelines are too broadly worded and complicated to apply reliably to individual beneficiaries. The SSA form that clinicians are asked to complete states the following: “Do you believe the patient is capable of managing or directing the management of benefits in his or her own best interest? By capable we mean that the patient: is able to understand and act on the ordinary affairs of life, such as providing for own adequate food, housing, clothing, etc., and is able, in spite of physical impairments, to manage funds or direct others how to manage them” (11).

There are ambiguities in these SSA guidelines. Differentiating individuals who are capable from those who are not requires subjective judgments about what it means to spend money in one’s best interest and how to direct others to manage funds. Given the broad guidelines provided by the SSA, it is not surprising that rates of assignment of representative payees vary widely across sites, which appears to reflect differences in assignment procedures rather than true individual differences in need (12,13).

Legal determinations of incapability are supposed to be based, first, on a functional assessment of skills and behaviors related to a beneficiary’s ability to make financial decisions and, second, on evidence that a person will suffer substantial harm from specific inabilities to manage finances or affairs (14). Surveyed clinicians report recommending payee assignment on the basis of clinical indicators, such as the client’s substance abuse or dependence, hospitalizations, homelessness, whether a beneficiary will accept a payee, and the effect that such a recommendation would have on the clinical relationship (1517).

In this study, financial capability was assessed for 118 adults with severe mental illness and histories of substance abuse who received and managed their own SSI or SSDI payments (18,19). To better understand challenges involved in making capability determinations, this article describes ten beneficiaries for whom capability determinations were not clear cut, even though a comprehensive assessment of financial capability was conducted.

Methods

Study Design

This study was approved by the Yale University Institutional Review Board. Altogether, 118 study participants were recruited between February 2011 and August 2012 from an inpatient psychiatric unit and two intensive outpatient programs in Connecticut. Eligible participants were between the ages of 18 and 65, spoke English, received at least $600 from SSI or SSDI each month, had a current or past DSM-IV substance use disorder diagnosis, did not have a representative payee or a conservator, and had a treating clinician who agreed to participate in the study.

Measure

Four independent assessors used the Financial Incapability Structured Clinical Assessment done Longitudinally (FISCAL) to measure financial incapability. The FISCAL is an assessment of financial capability with good interrater reliability and construct validity (18). [A table presenting reliability data is included in an online data supplement to this article.] It involves assessors using all data available to rate beneficiaries on whether the following five criteria apply: basic needs have not been met, funds needed for basic needs were spent on something else, the beneficiary spent substantial funds on something that harmed him or her, misspending was likely to continue, and there were contextual factors that would likely affect a beneficiary’s financial capability (Table 1). FISCAL assessors used an algorithm to make a final capability determination after considering and rating all five incapability criteria. [A flowchart depicting use of the algorithm with this sample is included in the online data supplement.]

Table 1 Financial capability criteria used in the Financial Incapability Structured Clinical Assessment done Longitudinally (FISCAL) to measure financial incapability

CriteriaHow to score
Criterion 1: basic needs have not been met: ?–inadequate information; 1, absent or false; 2, subthreshold; 3, threshold or trueBasic needs have not been met if client does not have stable housing, adequate food, or adequate clothing. If client does not spend funds to meet basic needs but is able to meet basic needs in some other way, this criterion for incapability is not met. It is possible to meet basic needs by planned use of food kitchens for food and public shelters for housing.
Criterion 2: funds needed for basic needs were spent on something else: ?–inadequate information; 1, absent or false; 2, subthreshold; 3, threshold or trueMoney needed for basic needs was spent on other things that are not essential. This criterion distinguishes between not meeting basic needs because of poverty, as opposed to not meeting basic needs because substantial funds were spent on nonessential items. Some clients don’t misspend their money; they just don’t have enough money no matter how it is spent. These clients are poor, not incapable.
Criterion 3: substantial funds spent on something that harmed the client: ?–inadequate information; 1, absent or false; 2, subthreshold; 3, threshold or trueClient’s funds were used for drugs, tobacco, gambling or other purchases that caused the client substantial harm. Substantial harm is defined by DSM criteria for substance abuse, which requires that substance use led to failure to fulfill major role obligations, recurrent use in situations that are hazardous, recurrent substance-related legal problems, or recurrent social problems exacerbated by substance use.
Criterion 4: past misspending (on harmful things or nonessentials) is likely to continue: ?–inadequate Information; 1, not likely; 2, as likely as not; 3, more likely than not; 4, not applicable (no past misspending)Likelihood that client’s misspending on harmful things or on nonessentials in the past 6 months will continue in the next year. Assessor must make judgment about whether new circumstances may affect past misspending on harmful things or on nonessentials. Past misspending on harmful things can be affected in several ways, for example through enrollment in a substance abuse treatment program to address substance abuse problems or attending counseling to treat a gambling addiction. Past misspending on nonessentials can be affected in several ways, for example a new medication regimen to control manic or impulsive symptoms or working with a money manager, conservator, or representative payee so that money is managed better.
Criterion 5: contextual factorsExamples: impaired judgment and disorganized behavior; someone else recommending or opposing payee assignment; physical limitations that compromise capability; new intervention that is likely to improve the beneficiary’s funds management

Table 1 Financial capability criteria used in the Financial Incapability Structured Clinical Assessment done Longitudinally (FISCAL) to measure financial incapability

Enlarge table

Data Sources Used in Fiscal Rating

Assessors used all available data from a comprehensive set of participant assessments conducted as part of a money management assessment study (20), clinical records, and a questionnaire about recent functioning and money management completed by a treating clinician (19). Assessors also conducted a semistructured clinical interview with each participant, inquiring about the participant’s expenditures, living situation, times in the previous six months when the participant did not have enough money for basic needs, how much money had been spent on harmful things, and plans for the future. For a separate analysis of FISCAL interrater agreement, one-third of the participants’ cases (N=44) were evaluated by a second independent assessor.

Identifying Ambiguous Cases

We defined cases as ambiguous when ratings by two independent assessors were discordant on any of the five FISCAL capability criteria or when assessors reported that an individual’s financial capability was difficult to judge.

Data Analysis

Cases that were identified as ambiguous were reviewed with assessors to identify factors accounting for the ambiguity in determinations.

Results

Of the 118 participants, ten were difficult to classify as financially capable or not. Demographic characteristics of these individuals (age, sex, and race-ethnicity) were similar to those of the full sample (18). [A table summarizing sample characteristics by financial capability is included in the online supplement.] The ten identified participants had a mean age of 44.6 years, and half were women. Six participants were Caucasian (non-Hispanic), three were Hispanic (Puerto Rican), and one was African American (non-Hispanic). Of the ten cases identified as ambiguous, five had discordant ratings on at least one of the incapability criteria and seven were identified as difficult to judge.

Sources of Ambiguity

Distinguishing incapability from the challenges of navigating poverty caused ambiguity.

For two participants, ambiguities arose because it was unclear whether poverty or nonessential spending had played a greater role in the participant’s failure to meet basic needs. One participant reported spending money on organic food, causing her to run short of money midway through the month. She also reported lending money to others despite not always having enough money to meet her own needs. Lack of funds contributed to her occasionally going hungry, as well as missing medical appointments because she was unable to pay for transportation. However the participant’s income was so small that even if she had not spent any money on nonessential items, she may still have had difficulty meeting her basic needs.

A second participant reported spending most of her income on essentials, but she occasionally spent money on things that she could not afford (for example, pets and loaning money to others). She reported difficulty paying bills and meeting basic needs. However, support from family and friends prevented her from losing her housing. In the recent past, she had gone hungry and lost weight after her food stamps were cut off.

The amount of nonessential spending that had to occur for a participant to be considered incapable contributed to ambiguity.

Ambiguities also arose in regard to the amount of nonessential spending when the beneficiary’s basic needs were being met through the help of outside resources and not by SSDI funds provided to the beneficiary for that purpose. One individual reported spending $350 per month on drugs and alcohol, $75 on dining out, and $100 on charitable donations. Most months, however, she was able to meet her basic needs with help from her husband’s income, money from her family, food stamps, and the occasional use of a food bank.

Another participant reported spending nearly half of her income on cigarettes and consequently ran low on food at the end of most months, could not replace her worn-out clothes, and purchased only medications that had no copays. Nevertheless, her needs were mostly met, and she was usually able to get a loan to cover her basic needs.

Modest spending on harmful things caused ambiguity.

For three beneficiaries, ambiguities were related to judgments about how much spending on harmful things renders someone incapable. In each case, the assessor had difficulty judging the participant’s financial capability because participants were spending only modest amounts, or nothing, on harmful things, but the consequences were often quite severe. Although substance use alone is not sufficient to find a person financially incapable (21), these beneficiaries’ substance use was associated with risky behaviors, vulnerability to victimization, and intoxication, all of which suggest that the beneficiaries were not acting in their own best interest, which may affect their ability to manage funds adequately. Assessors had to determine whether assigning a payee would likely ameliorate the negative consequences of substance use.

One participant spent only $60 a month on alcohol and received other drugs in exchange for letting people use his apartment. Even though the amount spent on alcohol was small, the participant’s alcohol use resulted in his discharge from methadone treatment, after which he relapsed to heroin use and had subsequent drug-related problems. Another participant reported receiving cocaine in return for helping drug dealers “run customers.” This participant had a long history of legal problems, hospitalizations, and social conflict associated with his drug use and was taking a large risk by working for drug dealers. A third participant spent an average of only $10 per month on alcohol but reported that she would occasionally binge drink, resulting in blackouts, hospitalizations, and legal problems.

Capability is fluid over time, which can create ambiguities.

The situations of two beneficiaries illustrate how financial capability is a fluid construct. Ambiguities arise depending on whether capability is assessed over a period of time or at one moment in time. In one case, a participant reported a significant period of time in the preceding six months during which he did not have enough money for food, and because he had recently been released from prison, he did not have a stable place to live. Subsequently, however, the participant started receiving food stamps and a few weeks later was able to find stable living arrangements. Looking at the six-month period as a whole, the participant was not meeting basic needs for most of the time, but at the time of the interview, the participant’s situation had stabilized and his basic needs were met. Another participant reported stable housing and utilities over the preceding six months but unstable medications, food, and clothing. Her needs were met for most of the six-month period, but episodic impulsive spending contributed to some financial hardship and unmet needs.

Predicting future stability caused ambiguity.

For four participants, ambiguities arose over the stability of supports that had helped a participant manage money. In one example, a participant would have failed to meet her basic needs from her Social Security payments without the intermittent help of her family and in-kind transfers from friends. At the time of the interview, the participant reported that she had asked her sister to help manage her affairs. The sister’s intervention was successful. However, because the participant had a history of rejecting help, the assessor felt it was unlikely that the participant would continue to allow her sister to assist and would continue to manage her funds poorly. In two other cases, a participant’s mother helped manage the participant’s finances, but there was inconsistent control of the funds and uncertainty about whether the beneficiaries would continue receiving help. In the case of a fourth beneficiary, the participant pooled resources with his roommate in a joint bank account. The roommate then paid all the bills. The participant was relatively unaware of his expenses, and the assessor had difficulty determining the stability of the roommate arrangement.

Discrepancies between sources of data (participant interviews, chart review, and clinician report) caused ambiguity.

Two capability determinations were ambiguous because of discrepancies between information collected from participant interviews, chart review, and clinician report. In both examples, the participants described themselves as more capable than was indicated in data from patient charts or from treating clinicians.

Discussion

Determining financial capability is complicated. One reason capability is difficult to judge is that managing a limited income, with or without a disabling illness, is very difficult. The challenges disabled people face—poverty, substance use (22), gambling (23), criminal victimization, financial dysfunction, psychiatric symptoms (24), and financial predation (6)—contribute to their financial difficulties. Most beneficiaries—and, in fact, most people—do not spend all of their funds on basic needs. A Bureau of Labor Statistics report found that Americans in the lowest, middle, and highest income quintiles spend 7%−10% of their income on nonessential items and that compared with those in the highest quintile, those in the lowest quintile spend a greater percentage of their money on basic necessities, such as housing, food, utilities, fuel for vehicles and public services, health care, and medications (25,26). Emerging literature suggests that because of the stresses of poverty, it is particularly difficult for someone who is poor to exert the planning, self-control, and attention needed to resist unnecessary purchases (27).

Second, determinations of the amount of nonessential or harmful spending and the circumstances around such spending that would merit payee assignment is a subjective judgment with few guidelines. The SSA guidelines about how representative payees must use a beneficiary’s monthly benefits allow for some nonessential purchases (that is, clothing and recreation) but only after food and shelter are provided for (28). This study highlighted areas requiring special deliberation. Clinicians assessing financial capability need to consider the extent to which spending patterns harm the individual being assessed. Misspending that results in a few missed meals might cause minor discomfort but not measurable harm, whereas misspending that results in an inability to pay for rent may be very harmful. When looking at harmful spending, clinicians should discern whether the beneficiary has a financial problem or an addiction problem. If improved financial skills or payee assignment would not have an impact on the acquisition of drugs of abuse, then the beneficiary’s substance use probably does not reflect financial incapability.

Another important issue that clinicians face when making determinations about beneficiaries’ ability to manage funds is attempting to predict future functioning, which is inherently uncertain. There is evidence that clinicians have difficulty predicting behaviors such as future medication adherence (29,30), and thus some uncertainty in predicting financial capability is to be expected. Frequent reevaluations of financial capability might help with complicated determinations. Extensive and serial evaluations of capability to manage one's funds are probably beyond the mandate and the resources of the SSA, but reevaluating the capability of beneficiaries who are admitted to an intensive treatment program may be feasible.

The issue of beneficiaries who may be incapable but who have unofficial arrangements with people who manage their funds was raised previously (31) and addressed in an audit by the Office of the Inspector General (OIG). Altogether, 13% of the SSI and SSDI beneficiaries evaluated by the OIG reported having people who received and managed their funds, even though these people had no formal role (1). Whether beneficiaries with unassigned surrogate money managers would benefit from more formal, monitored payee arrangements is uncertain (31). The impact of assigning a formal representative payee largely depends on the payee assigned, with wide variability in payee practices (32). The literature suggests that representative payee programs, particularly when coordinated with other psychiatric treatment, are beneficial (17,33,34).

Although applying standardized criteria can clearly identify most beneficiaries as capable or incapable of managing finances (19,20), in the absence of more precise guidance, capability determinations are left to clinicians' best judgment. In addition to considering whether a beneficiary is capable, clinicians should also consider whether assignment of a representative payee would be helpful (16) and whether payee assignment is feasible.

Conclusions

The ambiguous cases described in this article raise fundamental questions about how to determine financial incapability. Examining the details of peoples’ living situations and decision making can help in making dichotomous decisions about capability among the small proportion of beneficiaries for whom an algorithm leaves unresolved questions.

Ms. Lazar, Dr. Black, Dr. McMahon, and Dr. Rosen are with the Department of Psychiatry, Yale University School of Medicine, New Haven, Connecticut (e-mail: ). Ms. Lazar, Dr. Black, and Dr. Rosen are also with the Department of Veterans Affairs Connecticut Healthcare System, West Haven, Connecticut. Dr. O’Shea is with the Institute of Living, Hartford Hospital, Hartford, Connecticut.

This research was supported by grants R01DA025613 and R01DA12952 from the National Institutes of Health.

The authors report no financial relationships with commercial interests.

References

1 Disabled Individuals With Mental Impairments in Need of a Representative Payee. Washington, DC, Social Security Administration, Office of the Inspector General, 2012Google Scholar

2 Elbogen EB, Tiegreen J, Vaughan C, et al.: Money management, mental health, and psychiatric disability: a recovery-oriented model for improving financial skills. Psychiatric Rehabilitation Journal 34:223–231, 2011Crossref, MedlineGoogle Scholar

3 Marson DC, Savage R, Phillips J: Financial capacity in persons with schizophrenia and serious mental illness: clinical and research ethics aspects. Schizophrenia Bulletin 32:81–91, 2006Crossref, MedlineGoogle Scholar

4 Carpenter-Song E: Anthropological perspectives on money management: considerations for the design and implementation of interventions for substance abuse. American Journal of Drug and Alcohol AbuseGoogle Scholar

5 Rosen MI, Rosenheck RA, Shaner AL, et al.: Substance abuse and the need for money management assistance among psychiatric inpatients. Drug and Alcohol Dependence 67:331–334, 2002Crossref, MedlineGoogle Scholar

6 Claycomb M, Black AC, Wilber C, et al.: Financial victimization of adults with severe mental illness. Psychiatric Services 64:918–920, 2013LinkGoogle Scholar

7 Sells DJ, Rowe M, Fisk D, et al.: Violent victimization of persons with co-occurring psychiatric and substance use disorders. Psychiatric Services 54:1253–1257, 2003LinkGoogle Scholar

8 Shaner A, Eckman TA, Roberts LJ, et al.: Disability income, cocaine use, and repeated hospitalization among schizophrenic cocaine abusers—a government-sponsored revolving door? New England Journal of Medicine 333:777–783, 1995Crossref, MedlineGoogle Scholar

9 Appelbaum PS, Grisso T: The MacArthur Treatment Competence Study: I. mental illness and competence to consent to treatment. Law and Human Behavior 19:105–126, 1995Crossref, MedlineGoogle Scholar

10 Dunn LB, Jeste DV: Problem areas in the understanding of informed consent for research: study of middle-aged and older patients with psychotic disorders. Psychopharmacology 171:81–85, 2003Crossref, MedlineGoogle Scholar

11. Physician’s/Medical Officer’s Statement of Patient’s Capability to Manage Benefits. Form 787. Washington, DC, Social Security Administration. Available at www.ecpayee.org/uploads/8/8/7/4/8874367/doctor_statement_ssa_787.pdfGoogle Scholar

12 Rosen MI, McMahon TJ, Rosenheck RA: Homeless people whose self-reported SSI/DI status is inconsistent with Social Security Administration records. Social Security Bulletin 67:53–62, 2007MedlineGoogle Scholar

13 Appelbaum PS, Redlich A: Use of leverage over patients’ money to promote adherence to psychiatric treatment. Journal of Nervous and Mental Disease 194:294–302, 2006Crossref, MedlineGoogle Scholar

14 Guardianship: an agenda for reform. Mental and Physical Disability Law Reporter 13:271–313, 1989Google Scholar

15 Luchins DJ, Hanrahan P, Conrad KJ, et al.: An agency-based representative payee program and improved community tenure of persons with mental illness. Psychiatric Services 49:1218–1222, 1998LinkGoogle Scholar

16 Rosen MI, Ablondi K, Black AC, et al.: Pathways to assignment of payees. Community Mental Health Journal 50:270–274, 2014Crossref, MedlineGoogle Scholar

17 Hanrahan P, Luchins DJ, Savage C, et al.: Representative payee programs for persons with mental illness in Illinois. Psychiatric Services 53:190–194, 2002LinkGoogle Scholar

18 Lazar CM, Black AC, McMahon T, et al: FISCAL: An All-Data Approach to Assessing Financial Capability in People With Psychiatric Disabilities. New Haven, Conn, Yale School of Medicine, 2014. Available at www.behaviorchange.yale.edu/assess/363_190443_CPDDFISCALPoster2014final61014.pdfGoogle Scholar

19 Black AC, McMahon TJ, Rosenheck RA, et al.: Development of the Clinician Assessment of Financial Incapability (CAFI). Psychiatry Research 215:784–789, 2014Crossref, MedlineGoogle Scholar

20 Black RA, Rounsaville BJ, Rosenheck RA, et al.: Measuring money mismanagement among dually diagnosed clients. Journal of Nervous and Mental Disease 196:576–579, 2008Crossref, MedlineGoogle Scholar

21 Rosen MI, Rosenheck R: Substance use and assignment of representative payees. Psychiatric Services 50:95–98, 1999LinkGoogle Scholar

22 Results From the 2009 National Survey on Drug Use and Health: Mental Health Findings. Rockville, Md, Substance Abuse and Mental Health Services Administration, 2010Google Scholar

23 Petry NM, Stinson FS, Grant BF: Comorbidity of DSM-IV pathological gambling and other psychiatric disorders: results from the National Epidemiologic Survey on Alcohol and Related Conditions. Journal of Clinical Psychiatry 66:564–574, 2005Crossref, MedlineGoogle Scholar

24 Cook JA, Burke JK, Petersen CA: Assessing the Financial Planning Needs of Americans With Mental Illnesses. Chicago, University of Illinois at Chicago Center on Mental Health Services Research and Policy, 2004Google Scholar

25 Consumer Expenditure Survey. Washington, DC, US Department of Labor, Bureau of Labor Statistics, 2010Google Scholar

26 Edin K, Lein L: Making Ends Meet: How Single Mothers Survive Welfare and Low-Wage Work. New York, Russell Sage Foundation, 1997Google Scholar

27 Mullainathan S, Shafir S: Scarcity. New York, Henry Holt, 2013Google Scholar

28 A Guide for Representative Payees. Baltimore, Social Security Administration, 2009Google Scholar

29 Evans M: Clinicians often poor at predicting medication adherence. New York, ModernHealthcare, 2013. Available at www.modernhealthcare.com/article/20130521/NEWS/305219948/clinicians-often-poor-at-predicting-medication-adherence&template=mobileGoogle Scholar

30 Marcum ZA, Sevick MA, Handler SM: Medication nonadherence: a diagnosable and treatable medical condition. JAMA 309:2105–2106, 2013Crossref, MedlineGoogle Scholar

31 Elbogen EB, Swanson JW, Swartz MS, et al.: Characteristics of third-party money management for persons with psychiatric disabilities. Psychiatric Services 54:1136–1141, 2003LinkGoogle Scholar

32 Rosen MI, Bailey M, Dombrowski E, et al.: A comparison of satisfaction with clinician, family members/friends and attorneys as payees. Community Mental Health Journal 41:291–306, 2005Crossref, MedlineGoogle Scholar

33 Rosen MI: The “check effect” reconsidered. Addiction 106:1071–1077, 2011Crossref, MedlineGoogle Scholar

34 Conrad KJ, Lutz G, Matters MD, et al.: Randomized trial of psychiatric care with representative payeeship for persons with serious mental illness. Psychiatric Services 57:197–204, 2006LinkGoogle Scholar