The Evolution of Mental Health Policy and Economics
- Sherry GliedSherry GliedRobert F. Wagner Graduate School of Public Service, New York University
- and Richard FrankRichard FrankDepartment of Health Care Policy, Harvard Medical School
Mental health economics addresses problems that are common to all of health economics, but that occur with greater severity in this context. Several characteristics of mental health conditions—age of onset, chronicity, observability, and external effects—make them particularly economically challenging, and a range of policies have evolved to address these problems. The need for insurance—and for social insurance—to address mental health problems has grown. There is an expanding number of effective treatments available for mental health conditions, and these treatments can be relatively costly. The particular characteristics of mental health conditions exacerbate the usual problems of moral hazard, adverse selection, and agency. There is increased recognition, in both the policy and economics literatures, of the array of services and supports required to enable people with severe mental illnesses to function in society’s mainstream. The need for such non-medical services, generates economic problems of cross-system coordination and opportunism. Moreover, the impairments imposed by mental disorders have become more disruptive to the labor market because the nature of work is changing in a manner that creates special disadvantages to people with these conditions. New directions for mental health economics would address these effects.
Mental healthcare poses the same economic problems as other kinds of healthcare—only more so. Mental health conditions are prevalent, but severe illnesses (e.g., schizophrenia and bipolar disorder) are relatively rare. The etiology of mental health conditions includes genetic and environmental factors, and in almost all cases, the development and timing of these problems are unanticipated. Many mental health conditions are chronic relapsing illnesses and follow patterns similar to medical conditions like diabetes, congestive heart failure, and asthma. Common and serious mental disorders benefit from treatment, which is costly; the demand for treatment is responsive to prices. As with nearly all medical treatments, it is difficult for patients and third parties to assess the quality of care, and it is often difficult for providers or third parties to assess many subjective aspects of the patient’s condition. Mental illnesses interfere with work, family life, and social functioning. Together, these features that are shared with most health problems generate a demand for insurance to address the costs of treatment and the losses in income that result from illness. Because health insurance by its nature drives a wedge between the patient’s cost of illness and the total cost of treatment, moral hazard results. When insurance is allocated using market mechanisms, adverse selection threatens the efficient functioning of the insurance markets. Problems of adverse selection, as well as a desire to redistribute resources from those who are ex ante financially well off and healthy to those who ex ante have lower incomes and are not healthy, creates fundamental challenges for simple private insurance markets to efficiently offer protection against healthcare risks, at least for some groups. For example, private insurance markets are unlikely to provide the resources that would enable access to care over the lifetime for those who develop conditions early in life. In these cases, social insurance mechanisms may be preferred.
The standard economic challenges to health insurance markets, moral hazard and adverse selection, manifest in ways that are even more challenging to address in the mental health area, for a number of reasons, some of which are related to the conditions themselves. First, severe mental health problems commonly begin in adolescence and young adulthood, much younger ages than is the case for most other health conditions. This means that the adverse impacts on cognition and functioning stemming from mental health disorders occur during a period of investment in human capital and therefore have long-term effects on work, earnings, the development of social relations, and family. This differs markedly from conditions that occur later in life. Consequently, insurance for treatment of mental health conditions is likely to create relatively large distributive effects. Second, because the onset of the most disabling of mental illnesses occurs early in life, acute exacerbations will likely recur multiple times, for much longer than they do with other serious chronic illnesses. This means that their lifetime costs are greater and the incentives for insurers (and employers) to avoid people with mental disorders are especially strong. Those who develop serious mental health conditions before or just as they enter the labor market are unlikely to be able to save or to purchase health insurance coverage that will address their needs over the long term. Third, problems of observability—of the quality of treatments and the severity of underlying conditions—are more complex, because there is no “lesion” that characterizes a mental illness. Fourth, these problems may be more disruptive of families and communities than are most other medical conditions, in part because they affect younger people and last longer, but also because they affect social functioning and can create disruptions to individual relationships (family, coworkers, and friends) and to communities (violation of social norms). This means that a broad range of social service functions—including educational, public safety, health, housing, and employment—may be engaged in addressing these problems.
Insurance protects consumers against the uncertainty of economic losses—in this case, brought on by the development of a mental illness. In many countries, and particularly in the United States, health insurance protection against the risk of mental disorders is incomplete. Health insurance that pays for the treatment of mental health problems has often imposed greater barriers to accessing treatment for those illnesses than for other forms of care. Mental healthcare was historically often excluded from private insurance contracts and often was narrowly defined under public programs. Many of the income losses associated with mental illnesses—those that do not rise to the level of a qualifying disability—are either not insured at all or are incompletely insured through mainstream income and employment support programs (this is even truer of addictive disorders). Most disability insurance programs provide coverage at levels that are inadequate to support stable housing and access to supported employment. Moreover, the design of programs dampens incentives for people with mental illnesses to participate in the labor market at all. Finally, the organization of government social insurance typically represents a patchwork of programs that are uncoordinated and leave gaps.
Over time, the need for insurance—and for social insurance—to address mental health problems has grown. There are an expanding number of effective treatments available for mental health conditions, and the treatments can be relatively costly. There is new recognition of the array of services and supports required to successfully allow people with severe mental illnesses to function in society’s mainstream. The impairments imposed by mental disorders have become more disruptive to the labor market because the nature of work is changing in a manner that creates special disadvantages to people with these conditions.
Mental Illness and Its Consequences
In 2015, 18% of adults in the United States, or 43.4 million people, reported having any mental illness (AMI), a general term describing a range of mental, behavioral, or emotional disorders (U.S. Department of Health and Human Services, National Institute of Mental Health, 2015)—4.1% of adults met the criteria for a serious mental illness (SMI), defined by the National Institute of Mental Health (NIMH) as “a mental disorder with serious functional impairment which substantially interferes with or limits one or more major life activities” (Center for Behavioral Health Statistics and Quality, 2015; Kessler, Chiu, Demler, Merikangas, & Walters, 2005; U.S. Department of Health and Human Services, National Institute of Mental Health, 2015). Rates of illness in the United States are comparable to those in other high-income countries. Most mental disorders develop in adolescence and early adulthood, and this is also true of the most serious illnesses (Center for Behavioral Health Statistics and Quality, 2015; Kessler et al., 2007; Organisation for Economic Co-operation and Development [OECD], 2012). These illnesses then recur across the life span. It is estimated that about 50%–80% of current depression cases are recurrences of the illness (Burcusa & Iacono, 2007).
Because mental illnesses impair functioning, they impose significant costs on the affected individuals and the larger society. People with psychiatric disorders are 6 to 11 times more likely to be unemployed as are those without such disorders (Ettner, Frank, & Kessler, 1997; OECD, 2012). People with mental illnesses are also twice as likely as other people to experience long-term unemployment and to make claims on unemployment insurance and income support programs. They are also more likely to work in a low-skilled job or “a job with high psychological demand with low decision latitude” (OECD, 2012). One consequence is reflected in the fact that in the United States one out of four adults with a SMI lives below the poverty line (U.S. Department of Health and Human Services, Substance Abuse and Mental Health Services Administration, 2016). Likewise, the World Health Organization (WHO) reports that four of the ten leading causes of disability in high-income countries are mental disorders (National Alliance on Mental Illness, 2017b).
People with mental health problems often have cognitive and affective impairments that can make it difficult to form and maintain important personal and social relationships (Almedom, 2005; Ehsan & De Silva, 2015). Those impairments, therefore, can result in social isolation and reduced access to social capital (Webber et al., 2014). Failure to adhere to social norms and exhibiting disturbing behavior can further weaken social connections, which, in turn, can lead to worse health and more unhealthy behaviors (Umberson & Montez, 2010). The unpredictable behaviors of people with mental disorders can also contribute to stigmatization of these conditions, which may lead people to avoid treatment and diagnosis.
Because of their impairments in functioning, people with a SMI are disproportionately represented in the homeless population. It is estimated that 30% of the chronic homeless population have a SMI (Substance Abuse and Mental Health Services Administration [SAMHSA], 2015). They are also overrepresented among the incarcerated population. In the United States, an estimated 56% of those in state prisons and 64% of those in jails report symptoms of mental disorders: many are SMIs like schizophrenia and bipolar disorder. Moreover, 70% of individuals in this population with a mental disorder have a co-occurring substance use disorder (James & Glaze, 2006). Untreated and undertreated mental and substance use disorders raise the risk of reoffending and undermine the ability to adhere to the terms of probation or parole.
There is a body of evidence that establishes some correlation between mental illnesses and violent behavior (Tinelli & Kanavos, 2015). Prison studies have found that inmates with a mental disorder were more likely to have committed violent crimes than other inmates and have higher rates of prison misconduct and recidivism (Frank & McGuire, 2010; Kim, Becker-Cohen, & Serakos, 2015). Other studies have shown that children with antisocial behaviors are at risk of developing mental disorders as adults. People with mental disorders are also more likely than others to be victims of crime. The causal linkages between mental illnesses and crime are complex. The extant evidence indicates that people with serious mental illnesses who do not abuse substances are no more likely to engage in violent crimes than are otherwise similar people. However, people with a SMI are more likely to abuse substances than are similar people who do not suffer from mental illnesses. The co-occurrence of mental and substance use problems is associated with elevated levels of violence. It is this set of phenomena that underpins the correlation between mental illnesses and violence (Corrigan, Kerr, & Knudsen, 2005; Corrigan, Morris, Michaels, Rafacz, & Rüsch, 2012; Corrigan & Watson, 2002; Glied & Frank, 2016).
Adults living with a mental illness are more likely to die before age 65 than are otherwise similar individuals (Robert Wood Johnson Foundation, 2011). Typically, the deaths result from treatable chronic medical conditions. The proportion of people with comorbid mental health and physical disorders is about two to three times higher among adults 55–64 than among young adults (Center for Behavioral Health Statistics and Quality, 2015; OECD, 2012). It is, therefore, not surprising that Americans with serious mental illnesses also live shorter lives.
Mental disorders make large claims on economic resources. In the United States, for 2014, an estimated $219 billion was spent caring for mental and substance use disorders, of which $186 billion, or 85%, was spent on treating mental illnesses(SAMHSA, 2014). In addition to these direct costs of illness, the illnesses impose indirect costs related to functional impairment and social disruption previously discussed. The “economic burden” of individuals with a major depressive disorder increased by 21.5% (from $173.2 billion to $210.5 billion, inflation-adjusted dollars) between 2005 and 2010 (Greenberg, Fournier, Sisitsky, Pike, & Kessler, 2015). Fifty percent of the costs were attributed to workplace costs, while only 38% were attributed to direct medical costs (Greenberg et al., 2015). Employees with mental health problems incur significantly higher annual per capita costs and disability costs than other medical conditions and incur significantly more annual sick days (Druss, Rosenheck, & Sledge, 2000; OECD, 2015). Reduced productivity on the job, frequent sick days, and high healthcare costs contribute to significant workplace costs directly borne by employers (Wang et al., 2007; OECD, 2012). Kessler et al. found that depressed employees took more short-term disability days in a 30-day period than other workers. The disability days cost employers a salary-equivalent of between $182 and $395 over the 30-day period (Kessler et al., 1999).
Because mental illnesses create a range of functional impairments, many people with a serious mental illness rely on public income support, healthcare, and housing assistance programs. Mental disorders are the second leading diagnosis among Social Security Disability Insurance (SSDI) beneficiaries in the United States. Among people with a mental health disorder, the rise in disability rolls has been steep and coincident with a precipitous rise in unemployment in this population.
How Do We Protect Individuals and Communities From the Risks/Consequences of Mental Illnesses?
Mental illnesses have substantial costs. These include private costs of treatment and consequences of illness (unemployment, low earnings) on individuals and households, external social costs, such as those associated with crime and homelessness, and pecuniary impacts, such as claims on disability programs. As a consequence, all developed economies have adopted programs of private or social insurance to protect people from the financial risks associated with the development and treatment of these conditions.
In most countries, social protections against the risks of mental health problems were initially focused on those with the most serious mental illnesses, who were served under a system of largely government owned and operated psychiatric hospitals (Glied & Smith, 2013; Goldman & Grob, 2006). As health insurance and social insurance systems began to take shape, these structures evolved. In the United States, beginning in the mid-1960s, mainstream social insurance programs such as Medicaid and Medicare, SSI (Supplemental Security Income), and SSDI offered people with mental disorders broader access to insurance coverage and income support. In the decades since, the scope of covered conditions has grown, most recently through implementation of the Affordable Care Act, which expanded access to insurance coverage. Most Americans, including most of those with less disabling mental health problems, are covered by private health insurance, provided either by employers or purchased in the nongroup market. Until 2008, when Congress passed a law requiring “parity” in coverage, most private plans provided much more restricted insurance coverage for treatment of mental disorders than for other types of disorders.1 About 20% of plans sold in the nongroup insurance market excluded coverage for mental health disorders altogether. The Affordable Care Act required that coverage for mental health and substance use disorder treatment be considered essential health benefits and thus were required of all insurance sold in the individual and small group insurance markets. The result was to effectively require all health insurance to include mental health and substance use disorder coverage at “parity.”
A similar evolution has occurred with respect to social insurance against the effects of mental illness on functioning. In the United States, SSDI and SSI provide income support for people with a mental disorder significant enough to cause a long-term inability to engage in “gainful activity.” As of 2016, almost 9 million people received federally administered SSDI and SSI payments; 35.2% of them qualified for disability based on a mental disorder as defined by Social Security Administration (SSA; Social Security Administration Office of the Chief Actuary, 2017).
People who receive SSI and SSDI may qualify for federal and state housing assistance, including programs that are targeted at those with disabilities (U.S. Department of Housing and Urban Development, 2017). These subsidized properties are sometimes connected to supportive services, such as case management and employment assistance, usually through Medicaid (National Alliance on Mental Illness, 2017c). The availability of federal housing vouchers falls well below the estimated number of people who are chronically homeless and suffering from a mental disorder. Therefore, finding suitable housing can be difficult, and waitlists are long (Health Affairs, 2016).
A large body of literature shows evidence-based supported employment (SE) offers a cost-effective approach to engaging people with a SMI in the workforce (O’Day et al., 2014; OECD, 2015; Drake et al., 2013). Individual placement and support (IPS) programs are among the most effective supported employment programs (Campbell, Bond, & Drake, 2011; Mueser, Drake, & Bond, 2016). Studies have found that IPS programs have improved well-being, reduced psychiatric admissions (Hoffmann, Jackel, Glauser, Mueser, & Kupper, 2014), and helped formerly incarcerated individuals with mental disorders participate in the workforce (Mueser et al., 2016; Wang et al., 2007). Nevertheless, results from a large national randomized evaluation of IPS shows that it does not result in sufficient improvements in earnings capabilities to permit disability insurance (SSDI) recipients to exit the program (Frank, 2013). Most states do not support access to IPS. At the federal level, the Rehabilitation Services Administration (RSA) of the U.S. Department of Education, whose charge is workforce innovation, also does not fund IPS (National Alliance on Mental Illness, 2014) .
As in the United States, the economic burden of mental health disorders in most OECD countries is enormous. In the United Kingdom, mental health accounts for 4.5% of GDP (OECD, 2014). In Switzerland, 40% of people on disability have a mental health disorder. According to the OECD Mental Health and Work series, the lack of early identification and treatment is among the key challenges facing OECD countries. Just as in the United States, most policies in OECD countries target people once they are out of the workforce due to sickness or disability.
The structure of health and social insurance for people with mental disorders differs significantly across OECD countries (Glied & Smith, 2013). In decentralized healthcare systems that rely on private health insurance markets, like that of Switzerland, mental health is generally not an integrated part of the healthcare system. In more centralized public systems, like that in Sweden, mental health is an integrated part of the healthcare system, and provides healthcare, social services, and other in-kind benefits for people with serious disorders, but services do vary throughout the country (Commonwealth Fund, 2017). Increasingly, many OECD countries have made employment an explicit component of the treatment of mental illness.
Although the structure of health and social insurance programs for people with mental illness varies across OECD nations, several distinctive features of the programs are common. In general, insurance coverage for some forms of mental health treatment—particularly psychotherapy and long-term residential treatment—is provided with more restrictions than is typical of coverage for medical-surgical treatments. These may take the form of differences in benefit structure (whether outpatient psychotherapy provided by a psychologist is covered under health insurance, for example), differences in cost-sharing (e.g., the higher copayments required for mental health services prior to parity legislation in the United States), increased oversight and management of mental health services, or restrictions on supply.
A similar set of policies aimed at curtailing utilization is common in the design of programs to provide disability insurance and other social supports to people with mental health problems. All countries face the challenge of designing policies that provide adequate social support while also providing the incentives to limit reliance on these payments. One way is to make it more difficult to claim disability. Duggan and Imberman (2009) estimate that when the Social Security Administration in the United States made it easier for people with a mental illness to claim eligibility, they found that the “less strict criteria for mental health disorders” caused a 38% and 53% increase in disability among women and men, respectively (Duggan & Imberman, 2009). They offer little evidence about whether the baseline rate was “too low” or the increased level “too high.” Similarly, in 2010, the United Kingdom and the Netherlands reassessed eligibility for disability benefits. As a result of the overhaul, 27% and 40% of disability claimants (respectively) were found to be eligible for at least partial work.
In order to reduce perceived “over-claiming” at the front end, disability programs often limit benefits to those who cannot engage in gainful activity. This serves to reduce the incentive to leave disability insurance. In the United States, there are some special programs that make it possible for people with disabilities receiving SSDI or SSI to continue to work and maintain Medicare or Medicaid (Social Security Administration, 2017b). These special rules, however, apply only to people participating in certain performance-based work incentive programs (Social Security Administration, 2017a). Moreover, those programs are designed in a way that creates strong incentives for employment support agencies to not enroll people with mental disorders as clients, thereby reducing the potential gains from such programs (Frank & McGuire, 2003).
Why Is Designing Insurance So Hard?
Designing insurance for the consequences of mental health problems is difficult because the problems of moral hazard, adverse selection, and agency that are the focus of the general health economics literature are all more severe in mental health.
Insurance reduces the price paid for care and in so doing leads people to consume more medical care than they would have consumed otherwise (ex post moral hazard). The price elasticity of demand and the generosity of coverage are the two factors that determine the extent of the welfare loss from moral hazard. If the price elasticity of demand is low (inelastic), the welfare loss due to moral hazard is attenuated. There is evidence that under pure fee-for-service reimbursement arrangements (no supply-side rationing) demand for certain mental health services is more responsive to changes in prices than are other types of medical care. Observational, quasi-experimental, and experimental studies have shown that the price elasticity of demand for outpatient psychotherapy is more responsive to insurance than the demand for general medical services (Frank & McGuire, 1986). Early observational studies exploited existing variation in the generosity of health insurance contracts and consistently found that the elasticity of demand for mental health services (general psychotherapy and inpatient treatment) was greater than that for physical health services (Horgan, 1986; McGuire, 1981; Taube, Kessler, & Burns, 1986; Watts, Scheffler, & Jewell, 1986). Analysis of the RAND Health Insurance Experiment (HIE) showed an arc elasticity of coinsurance of -0.80 for psychotherapy compared to -0.30 for ambulatory health services (Keeler, Manning, & Wells, 1988; Newhouse & Insurance Experiment Group, 1993). Across a broad range of studies, a general conclusion is that under pure fee-for-service payment arrangements, the elasticity of demand for psychotherapy is twice that for general office visits (Frank & McGuire, 1986; Kaplan, Spittel, & David, 2015). Other studies have also found high price elasticity of demand for certain inpatient services, particularly adolescent residential treatment (Frank, Salkever, Sharfstein, 1991; Hutchins, Frank, & Glied, 2011).
In contrast, it is notable that theory and empirical evidence suggest that more severe or complex mental health disorders are unlikely to generate the same moral hazard effects (Aron-Dine, Einav, Finkelstein, & Cullen, 2012). This is likely true for two reasons. The first is that the private costs of having a SMI are so high that ex post moral hazard effects would be unlikely. The other is that people with a SMI do not usually determine the course of their own treatment, though they increasingly have more autonomy to do so. Instead, a provider is likely to determine the duration and amount of treatment for an individual with SMI (Frank & McGuire, 1986).
The high price elasticity of demand for psychotherapy explains why insurance coverage for mental health has historically been much less complete than for physical health, whether because of cost-sharing or benefit coverage rules (Frank, McGuire, & Newhouse, 1995). In the United States, this pattern began to change in the 1990s when health insurers developed a new set of care management tools. Using administrative oversight and network design, insurers were able to triage patients to appropriate levels of care without relying on cost-sharing (Lu, Frank, & McGuire, 2008). The evidence from a range of evaluations of the impact of managed care on utilization of mental health services showed that the demand responses from expanding insurance could be attenuated (Frank & Garfield, 2007). Managed care demonstrated to policymakers that mental health services could be expanded without commensurate increases in spending (Beronio, Glied, & Frank, 2014).
In the United States, the policy response to the changing rationing arrangements in mental health delivery was to enact the Mental Health Parity and Addictions Equity Act (MHPAE) of 2008, which required that insurers provide coverage for mental health and substance use disorder care on par with medical surgical coverage. Several recent studies have found that, after parity, there were only modest increases in utilization of and spending on mental health services (Antonisse, Garfield, Rudowitz, & Artiga, 2016; Busch et al., 2014; Ettner et al., 2016).
Mental health conditions are often chronic recurring conditions. Moreover, they are commonly comorbid with medical conditions, such as diabetes, hypertension, and congestive heart failure. In some cases, this is a consequence of the treatments for serious mental illnesses. In other cases, the medical condition can trigger psychiatric conditions (e.g., depression). This means that competing health plans or providers paid on a capitated basis have strong incentives to avoid patients with these conditions (Ellis & McGuire, 2007; Frank, Glazer, & McGuire, 2000). These incentives have been a central factor in limited mental health coverage even in the presence of managed care arrangements. More recently, public and private policymakers have crafted responses to address the incentives stemming from adverse selection.
The main institutional response to adverse selection in mental health services has been the adoption of carve-out contracts that, in competitive health insurance markets, span choices of health insurance and delegate all management of mental health and addiction services to a single organization. Under such carve-out contracts, a single insurer organization takes on all the risk of mental health. This type of contract removes the design of the mental health benefit from the insurer choices and market dynamics that create selection incentives. Instead, this arrangement makes mental health coverage a contractual matter between the payer and the managed behavioral healthcare organization.
Regulatory responses to adverse selection around mental health have taken two forms. First, regulators require insurers to cover certain mental health benefits, and they may also define the terms under which these benefits must be covered. During the 1980s and 1990s, dozens of states put into place minimum benefit statutes that required a specific level of mental health coverage (Frank & McGuire, 2000). An extension of this approach is the defining of mental healthcare as an essential health benefit and requiring parity in benefits, as was done under the Affordable Care Act. Nevertheless, through the use of administrative mechanisms, managed care plans can pursue incentives to compete to avoid enrollees who are persistently high cost. Second, regulators may use systems of risk adjustment or reinsurance to diminish the incentives for insurers to compete by avoiding those with mental health problems. The growing use of risk adjustment has led to a substantial literature on risk adjustment for mental health services (U.S. Department of Health and Human Services, Agency for Healthcare Research and Quality, 2015). Systems of risk adjustment can attenuate selection incentives, but risk adjustment systems in current use continue to leave in place strong incentives to avoid people with these conditions (Ettner, Frank, & Kessler, 1997; McGuire, Newhouse, Normand, Shi, & Zuvekas, 2014; Montz et al., 2016). These incentives lead competing insurance plans to underprovide welfare-improving mental health services, or to reduce the quality of services to discourage enrollment (Deb, Wilcox-Gök, Holmes, & Rubin, 1996; Padgett, Patrick, Burns, Schlesinger, & Cohen, 1993; Perneger, Allaz, Etter, & Rougemont, 1995; Reed, 1974; Rogers, Wells, Meredith, Sturm, & Burnam, 1993). While risk adjustment can aid in addressing adverse selection incentives, it comes at the cost of increased incentives to “upcode,” that is, the tendency of health plans to assign diagnoses to individual patients indicating greater complexity of illness than occurs for similar patients paid under fee-for-service arrangements.
Adverse selection is likely to be particularly serious with respect to the illnesses with the earliest age of onset (childhood mental illnesses that persist into adulthood). Health plans will have an incentive to adopt benefit designs and administrative mechanisms that make their plans unattractive to households with children suffering from mental and emotional problems.
Challenges to Insurance Design—Provider Payment
Mental health conditions are also unusually likely to generate agency problems. That is, insurers and providers may be able to exploit information asymmetries and patient impairments in order to pursue economic gains at the expense of the care provided to people with mental illnesses. Mental health conditions are often not readily observable or diagnosable, people with these conditions may have impairments that affect their ability to assess treatment quality, and many mental health treatments are very challenging to monitor (Hepner, Greenwood, Azocar, Miranda, & Burnam, 2010). Moreover, stigma may make patients reluctant to admit to themselves or others that they would benefit from mental health treatments. Together these forces have contributed to public and private payers’ being reluctant to use “high-powered” financial incentives, like capitated payment mechanisms, to pay for mental healthcare, as they often do for general healthcare (Lave, 2003).
Improvements in market and policy design have substantially dampened (though not eliminated) the market failures in mental health insurance that result from moral hazard, adverse selection, and agency. Instead, several new frontiers of health economics research have emerged. These include program alignment and coordination of care across diverse public programs and service providers; promoting early engagement (first episode) with treatment, especially for those with the most serious disorders (schizophrenia and bipolar disorders); and the challenges of developing social insurance models that encompass treatment, employment/disability, and social inclusion.
Providing insurance for the cost of mental health treatments supplied by conventional providers (doctors, hospitals) is not enough to mitigate the underlying risk associated with mental health conditions. People with SMIs need to be insured against a range of needs that span health and social service sectors. In fragmented systems of social service delivery, this frequently leads to coordination problems, and “wrong-pocket problems” of cross-system externalities. For example, cutting back on residential care for people with serious disorders in the mental health system may generate a need for housing that falls on the public housing system or may result in higher levels of incarceration. The earliest studies of this type focused on the period of deinstitutionalization, when financial incentives coupled with treatment changes led to substantial declines in inpatient psychiatric hospitalization. As the use of inpatient residential services declined, however, utilization increased in nursing homes and other residential treatment facilities (Goldman, Adams, & Taube, 1983).
One set of studies has shown that managed mental healthcare programs may shift costs to the general health sector (Mechanic, Schlesinger, & McAlpine, 1995). Many studies have shown that mainstream services and social supports cost-shift to jails, prisons, and emergency rooms. Cost-shifting occurs when the government underpays expected costs and shifts the economic burden to another entity that serves the same populations. A recent quasi-experimental study found evidence that such cost-shifting can occur through changes in care management arrangements as well as payment mechanisms. The study compared Medicaid enrollees and non-Medicaid enrollees before and after the introduction of managed care and found that Medicaid beneficiaries experience a significant increase in the likelihood of being incarcerated following the shift to managed care compared to the groups where no changes in care rationing were made (Domino, Norton, Morrissey, & Thakur, 2004).
The reverse side of cost-shifting is the problem of care coordination across agencies. Governmental agencies are not always coordinated to provide care for the underlying health-related social needs of people with mental disorders (U.S. Department of Health & Human Services, Office of the Assistant Secretary for Planning and Evaluation, 2014). Evidence-based programs, such as rehabilitation services, assertive community treatment (ACT), IPS, and services provided by community support teams, require coordination across a variety of otherwise independent public institutions and authorities. Coordinating care for people with disabilities and complex health needs is beset by unwieldy administrative challenges and complex payment mechanisms. For instance, in some states because substance abuse treatment is not defined as a mental health benefit, substance abuse treatment cannot be reimbursed without a diagnosis of a mental disorder (U.S. Department of Health & Human Services, Office of the Assistant Secretary for Planning and Evaluation, 2014). Coordinating care is also a challenge, as health insurance programs do not always pay for administrative costs, such as case management (U.S. Department of Health & Human Services, Office of the Assistant Secretary for Planning and Evaluation, 2014). In the United States, health policy has tried to address this through the introduction of broader reimbursement arrangements, such as Health Homes, which aim to coordinate care for health and human services, and Accountable Health Communities, which are intended to improve coordination across communities (Alley, Asomugha, Conway, & Sanghavi, 2016).
Because serious mental illnesses impose consequences that extend beyond the affected person, society may wish to do more than facilitate access to mental health services. In extreme cases, it may compel use of these services (through civil commitment). In cases where violence to oneself or others is not an important consideration, society may, nonetheless, wish to promote access to care where this might prevent future disability or family dissolution. One important opportunity where treatment science has developed new interventions that offer promising results for altering the trajectory of impairment disability is in the area of reducing the duration of untreated psychosis. Recent services research shows that programs that target early-onset patients reduce the average duration of untreated psychosis for new patients with schizophrenia by 2.5 to 4 years. Research findings from large- and small-scale trials of interventions aimed at treating first episodes of psychoses support the view that engaging people with psychoses in care early can stem deterioration in functioning. The findings show greater effectiveness and much greater cost-effectiveness for people who are engaged in treatment within the first year of their illness. These models involve some services that fit comfortably into existing insurance arrangements (physician visits, medications) but also services not routinely covered by all health insurance programs (case management, family psycho-education, supported employment/education). In addition, the models call for outreach activities that are traditionally associated with public health interventions like vaccinations. This broader and more diverse package of services does not naturally fit into most existing insurance programs. Nor does it fit into programs targeted at people with disabilities, because the programs seek to avert disabilities before they occur. The challenge going forward is to design a financing mechanism that leverages the program elements covered by insurance and allows for getting the payment incentives “right.” The payment system and financing mechanisms need to be designed to balance incentives for case finding, concerns about efficient supply of care, and rewarding high-quality care. These multiple objectives suggest the use of multipart payment mechanisms that combine case rates, fee-for-service, and performance-based contracting features (Frank, McGuire, & Glied, 2015).
While significant progress has been made on inclusion of mental health in health insurance systems, there is clear evidence that people with mental illnesses are increasingly disconnected from work and social engagement. Across the developed world, the number of people disabled because of a mental health condition is rising. This increase is occurring despite evidence that mental disorders are no more common today than they were in the past 50 years (OECD, 2012) and despite an increase over time in the percentage of people with a mental illness receiving effective treatment (Frank & Glied, 2006).
One potential explanation of the rise in disability among people with mental illnesses relates to moral hazard with respect to income support programs. The combination of stagnant wage growth for lower-income workers and economic inequality has led to rising “replacement rates” (the fraction of one’s income that can be replaced with disability benefits or other income support programs) and that produces moral hazard. Eligibility for a disability program may lead marginal workers to stop working and claim disability instead (Black, Daniel, & Sanders, 2002; Maestas, Mullen, & Strand, 2013; Moore, 2015). The eligibility determination process for SSDI and SSI in the United States is long and cumbersome and requires a high degree of impairment. Nevertheless, the disability benefit system is often seen as a safety net of last resort, and access to it is based on a test of the ability to engage in substantial gainful activities. Moreover, moral hazard may lead people to remain on disability programs. These programs give people financial disincentives to return to work (Rosenheck et al., 2006).
However, rising replacement rates and moral hazard offer at best a partial explanation of the decline in employment rates and the upward trend in disability applications. A different force affecting these patterns may be that the opportunities for people with mental illnesses to engage in substantial gainful activities are shrinking. Acemoglu and Autor offer evidence suggesting that the nature of work is changing (Acemoglu & Autor, 2011). The mechanisms shaping employment and disability trends in the United States and other high-income economies are the “interactions among worker skills, job tasks, evolving technologies, and shifting trading opportunities.” Jobs requiring “mental resilience,” “situation adaptability,” and “communications ability” are on the rise. Future jobs are likely to involve tasks that are difficult to automate. These include abstract tasks that require “problem-solving capabilities, intuition, creativity, and persuasion,” and interpersonal tasks that “require situational adaptability, visual and language recognition, and in-person interactions” (Acemoglu & Autor, 2011, p. 1077). These abstract and interpersonal tasks require the cognitive and social skills that prove most challenging for people with mental health disorders. The changing nature of work could be driving the premature exit of low-skilled workers from the labor force.
There are notable implications of these observations for economic analysis and policy design. There is a body of research on clinical interventions targeted at “cognitive remediation”—interventions that are aimed at improved cognitive functioning, such as memory, planning, and cognitive flexibility (Cicerone et al., 2011). While a host of clinical trials show the benefits of these interventions, they have yet to be translated into programs that have been shown to strongly affect work and earnings. In addition, people with SMI will likely require targeted job training and specialized accommodations. These treatment and jobs programs, in turn, will require new models of coverage and new payment structures, to extend the scope of protections offered by insurance.
Similarly, economics has yet to grapple with the problem of engaging people with treatment early in order to prevent or at least delay the onset of disability. This will require reconciling the economics of insurance-related moral hazard and the economics of health externalities. It will require a program of research that identifies the most cost-effective techniques for identification and engagement of people early in their course of illness and designing an accompanying payment system.
Mental health economics is grounded in the characteristics of mental illness and treatment and in the specifics of mental health policy. An improved understanding of the nature of mental illness and its treatment has led policymakers to integrate people with mental health problems into mainstream society, so that mental health policy and mental health economics are no longer primarily about the funding of institutionalized care. In response to this interest in including people with mental illness in the broader society, mental health policy development and the field of mental health economics research have grown together. For example, the expanded understanding of the economics of insurance for mental healthcare has informed policy design, and the implementation of those policies has resulted in improved protection against the financial risks of treating mental illnesses. Moreover, because problems of mental health economics are often thornier cases of more general health economics problems, research in mental health economics has also contributed to health policy economics, particularly around issues such as benefit design and risk adjustment. Changes in policy around insurance for mental health treatment, often drawing on mental health economics research, have had very substantial, and positive, effects on the lives of people with mental illness (Frank & Glied, 2006).
A second wave of mental health economic research has begun to focus on problems and policy solutions around providing insurance against the non-medical-treatment consequences of mental illness. This research examines how to provide and finance services and supports that extend beyond, but are often coordinated with, those offered by the healthcare sector. These studies have raised issues around coordination of care across diverse public agencies and problems of cost-shifting. Here, too, insights from studies of mental health economics and policy have contributed to economics and policy understanding of programs aimed at high utilizers in other areas of health.
Evidence of rising disability rates and poor labor market functioning now suggests a third focus for research in mental health economics, aligned with the policy interest in the social inclusion of people with mental health conditions. This research needs to focus on how to connect people to treatment before they suffer significant impairments, and how to address the mismatch between functional capabilities and the changing requirements of the labor market. Once again, the substantial and immediate problems of people with serious mental illness likely presage problems that will affect broader populations as well. A better understanding of how to use economics and policy to build social inclusion in this sphere will be critical to improving the well-being of people with mental illness, and is likely to be valuable, as well, in thinking about problems of labor market mismatch more generally.
We thank Anais Borja for her excellent assistance in researching this article.
- Acemoglu, D., & Autor, D. (2011). Skills, tasks and technologies: Implications for employment and earnings. Handbook of Labor Economics, 4, 1043–1171.
- Alley, D. E., Asomugha, C. N., Conway, P. H., & Sanghavi, D. M. (2016). Accountable health communities—Addressing social needs through Medicare and Medicaid. New England Journal of Medicine, 374(1), 8–11.
- Almedom, A. M. (2005). Social capital and mental health: An interdisciplinary review of primary evidence. Social Science & Medicine, 61, 943–964.
- Antonisse, L., Garfield, R., Rudowitz, R., & Artiga, S. (2016). The effects of Medicaid expansion under the ACA: Findings from a literature review. http://www.kff.org/report-section/the-effects-of-medicaid-expansion-under-the-aca-findings-from-a-literature-review-issue-brief/
- Aron-Dine, A., Einav, L., Finkelstein, A., & Cullen, M. R. (2012). Moral hazard in health insurance: How important is forward looking behavior? Cambridge, MA: National Bureau of Economic Research.
- Beronio, K., Glied, S., & Frank, R. (2014). How the Affordable Care Act and Mental Health Parity and Addiction Equity Act greatly expand coverage of behavioral health care. Journal of Behavioral Health Services & Research, 41(4), 410–428.
- Black, D., Daniel, K., & Sanders, S. (2002). The impact of economic conditions on participation in disability programs: Evidence from the coal boom and bust. American Economic Review, 92(1), 27–50.
- Burcusa, S. L., & Iacono, W. G. (2007). Risk for recurrence in depression. Clinical Psychology Review, 27(8), 959–985.
- Busch, S. H., Epstein, A. J., Harhay, M. O., Fiellin, D. A., Un, H., Leader, D., Jr., & Barry, C. L. (2014). The effects of federal parity on substance use disorder treatment. American Journal of Managed Care, 20(1), 76–82.
- Campbell, K., Bond, G. R., & Drake, R. E. (2011). Who benefits from supported employment: A meta-analytic study. Schizophrenia Bulletin, 37(2), 370–380.
- Center for Behavioral Health Statistics and Quality. (2015). Behavioral health trends in the United States: Results from the 2014 National Survey on Drug Use and Health. HHS publication no. SMA 15–4927, NSDUH Series H-50.
- Cicerone, K. D., Langenbahn, D. M., Braden, C., Malec, J. F., Kalmar, K., Fraas, M., . . . Ashman, T. (2011). Evidence-based cognitive rehabilitation: Updated review of the literature from 2003 through 2008. Archives of Physical Medicine and Rehabilitation, 92(4), 519–530.
- Commonwealth Fund. (2017). The Swedish health care system. International Health Care System Profiles. http://international.commonwealthfund.org/countries/sweden/
- Corrigan, P. W., Kerr, A., & Knudsen, L. (2005). The stigma of mental illness: Explanatory models and methods for change. Applied and Preventive Psychology, 11, 179–190.
- Corrigan, P. W., Morris, S. B., Michaels, P. J., Rafacz, J. D., & Rüsch, N. (2012). Challenging the public stigma of mental illness: A meta-analysis of outcome studies. Psychiatric Services, 63(10), 963–973.
- Corrigan, P. W., & Watson, A. C. (2002). Understanding the impact of stigma on people with mental illness. World Psychiatry, 1(1), 16–20.
- David, H. (2015). Why are there still so many jobs? The history and future of workplace automation. Journal of Economic Perspectives, 29(3), 3–30.
- Deb, P., Wilcox-Gök, V., Holmes, A., & Rubin, J. (1996). Choice of health insurance by families of the mentally ill. Health Economics, 5(1), 61–76.
- Domino, M. E., Norton, E. C., Morrissey, J. P., & Thakur, N. (2004). Cost shifting to jails after a change to managed mental health care. Health Services Research, 39(5), 1379–1401.
- Drake, R. E., Frey, W., Bond, G. R., Goldman, H. H., Salkever, D., Miller, A., . . . Milfor, R. (2013). Assisting Social Security Disability Insurance beneficiaries with schizophrenia, bipolar disorder, or major depression in returning to work. American Journal of Psychiatry, 170(12), 1433–1441.
- Druss, B. G., Rosenheck, R. A., & Sledge, W. H. (2000). Health and disability costs of depressive illness in a major U.S. corporation. Primary Care Companion to the Journal of Clinical Psychiatry, 2(5), 185.
- Duggan, M., & Imberman, S. A. (2009). Why are the disability rolls skyrocketing? The contribution of population characteristics, economic conditions, and program generosity. In D. Cutler & D. Wise (Eds.), Health at older ages: The causes and consequences of declining disability among the elderly (pp. 337–379). Chicago, IL: University of Chicago Press.
- Ehsan, A. M., & De Silva, M. J. (2015). Social capital and common mental disorder: A systematic review. Journal of Epidemiology and Community Health, 69(10), 1021–1028.
- Ellis, R. P., & McGuire, T. G. (2007). Predictability and predictiveness in health care spending. Journal of Health Economics, 26, 25–48.
- Ettner, S. L., Harwood, M., Thalmayer, J. A., Ong, M. K., Xu, H., Bresolin, M. J., . . . Azocar, F. (2016). The Mental Health Parity and Addiction Equity Act evaluation study: Impact on specialty behavioral health utilization and expenditures among “carve-out” enrollees. Journal of Health Economics, 50, 131–143.
- Ettner, S. L., Frank, R. G., & Kessler, R. C. (1997). The impact of psychiatric disorders on labor market outcomes. ILR Review, 51(1), 64–81.
- Frank, R. G. (2013). Supported employment: Evidence on economic impacts. Psychiatric Services, 64(2), 103.
- Frank, R. G., & Garfield, R. L. (2007). Managed behavioral health care carve-outs: Past performance and future prospects. Annual Review of Public Health, 28, 303–320.
- Frank, R. G., Glazer, J., & McGuire, T. G. (2000). Measuring adverse selection in managed health care. Journal of Health Economics, 19(6), 829–854.
- Frank, R. G., & Glied, S. (2006). Better but not well: Mental health policy in the United States since 1950. Baltimore, MD: Johns Hopkins University Press.
- Frank, R. G., & McGuire, T. G. (1986). A review of studies of the impact of insurance on the demand and utilization of specialty mental health services. Health Services Research, 21(2 Pt 2), 241–265.
- Frank, R. G., & McGuire, T. G. (2000). Economics and mental health. In A. J. Culyer & J. P. Newhouse (Eds.), Handbook of health economics (Vol. 1, Part B, pp. 893–954). Amsterdam, The Netherlands: Elsevier.
- Frank, R. G., & McGuire, T. G. (2003). Setting payments in the Ticket to Work Program: Applying experience from capitation payments in health insurance. In K. Rupp & S. H. Bell (Eds.), Paying for results in vocational rehabilitation: Will provider incentives work for Ticket to Work? Washington, DC: Urban Institute.
- Frank, R. G., & McGuire, T. G. (2010). Mental health treatment and criminal justice outcomes. In P. J. Cook, J. Ludwig, & J. McCrary (Eds.), Controlling crime: Strategies and tradeoffs (pp. 167–207). Chicago, IL: University of Chicago Press.
- Frank, R., McGuire, T., & Glied, S. (2015). Paying for early interventions in psychoses: A three-part model. Psychiatric Services, 66(7), 677–679.
- Frank, R. G., McGuire, T. G., & Newhouse, J. P. (1995). Risk contracts in managed mental health care. Health Affairs (Project Hope), 14(3), 50–64.
- Frank, R. G., Salkever, D. S., & Sharfstein, S. S. (1991). A new look at rising mental health insurance costs. Health Affairs, 10(2), 116–123.
- Glied, S., & Frank, R. G. (2016). Economics and the transformation of the mental health system. Journal of Health Politics, Policy and Law, 41(4), 541–558.
- Glied, S., & Smith, P. (2013). The Oxford handbook of health economics. Oxford, UK: Oxford University Press.
- Goldman, H. H., Adams, N. H., & Taube, C. A. (1983). Deinstitutionalization: The data demythologized. Psychiatric Services, 34(2), 129–134.
- Goldman, H. H., & Grob, G. N. (2006). Defining “mental illness” in mental health policy. Health Affairs, 25(3), 737–749.
- Greenberg, P. E., Fournier, A.–A., Sisitsky, T., Pike, C. T., & Kessler, R. C. (2015). The economic burden of adults with major depressive disorder in the United States (2005 and 2010). Journal of Clinical Psychiatry, 76(2), 155–162.
- Health Affairs. (2016). Health policy brief: Medicaid and permanent supportive housing. online.
- Hepner, K., Greenwood, G., Azocar, F., Miranda, J., & Burnam, M. (2010). Usual care psychotherapy for depression in a large managed behavioral health organization. Administration and Policy in Mental Health and Mental Health Services Research, 37(3), 270–278.
- Hoffmann, H., Jackel, D., Glauser, S., Mueser, K. T., & Kupper, Z. (2014). Long-term effectiveness of supported employment: 5-year follow-up of a randomized controlled trial. American Journal of Psychiatry, 171(11), 1183–1190.
- Horgan, C. M. (1986). The demand for ambulatory mental health services from specialty providers. Health Services Research, 21(2), 291–319.
- Hutchins, E. C., Frank, R. G., & Glied, S. A. (2011). The evolving private psychiatric inpatient market. Journal of Behavioral Health Services & Research, 38(1), 122–131.
- James, D. J., & Glaze, L. E. (2006). Mental health problems of prison and jail inmates. U.S. Department of Justice, Office of Justice Programs. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.694.3355&rep=rep1&type=pdf
- Kaplan, R. M., Spittel, M. L., & David, D. H. (2015). Population health: Behavioral and social science insights. Washington, DC: Government Printing Office.
- Keeler, E. B., Manning, W. G., & Wells, K. B. (1988). The demand for episodes of mental health services. Journal of Health Economics, 7(4), 369–392.
- Kessler, R. C., Amminger, G. P., Aguilar-Gaxiola, S., Alonso, J., Lee, S., & Ustun, T. B. (2007). Age of onset of mental disorders: A review of recent literature. Current Opinion in Psychiatry, 20(4), 359–364.
- Kessler, R. C., Barber, C., Birnbaum, H. G., Frank, R. G., Greenberg, P. E., Rose, R. M., et al. (1999). Depression in the workplace: Effects on short-term disability. Health Affairs (Project Hope), 18(5), 163–171.
- Kessler, R. C., Chiu, W. T., Demler, O., Merikangas, K. R., & Walters, E. E. (2005). Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey Replication. Archives of General Psychiatry, 62(6), 617–627.
- Kim, K., Becker-Cohen, M., & Serakos, M. (2015). The processing and treatment of mentally ill persons in the criminal justice system. Washington, DC: The Urban Institute.
- Lave, J. R. (2003). Developing a Medicare prospective payment system for inpatient psychiatric care. Health Affairs, 22(5), 97–109.
- Lu, C., Frank, R. G., & McGuire, T. G. (2008). Demand response of mental health services to cost sharing under managed care. Journal of Mental Health Policy and Economics, 11(3), 113–125.
- Maestas, N., Mullen, K. J., & Strand, A. (2013). Does disability insurance receipt discourage work? Using examiner assignment to estimate causal effects of SSDI receipt. American Economic Review, 103(5), 1797–1829.
- McGuire, T. G. (1981). Financing and demand for mental health services. Journal of Human Resources, 16(4), 501–522.
- McGuire, T. G., Newhouse, J. P., Normand, S.-L., Shi, J., & Zuvekas, S. (2014). Assessing incentives for service-level selection in private health insurance exchanges. Journal of Health Economics, 35, 47–63.
- Mechanic, D., Schlesinger, M., & McAlpine, D. D. (1995). Management of mental health and substance abuse services: State of the art and early results. Milbank Quarterly, 73(1), 19–55.
- Montz, E., Layton, T., Busch, A. B., Ellis, R. P., Rose, S., & McGuire, T. G. (2016). Risk-adjustment simulation: Plans may have incentives to distort mental health and substance use coverage. Health Affairs, 35(6), 1022–1028.
- Moore, T. J. (2015). The employment effects of terminating disability benefits. Journal of Public Economics, 124, 30–43.
- Mueser, K. T., Drake, R. E., & Bond, G. R. (2016). Recent advances in supported employment for people with serious mental illness. Current Opinion in Psychiatry, 29(3), 196–201.
- National Alliance on Mental Illness. (2014). Road to recovery: Employment and mental illness. https://www.nami.org/About-NAMI/Publications-Reports/Public-Policy-Reports/RoadtoRecovery.pdf
- National Alliance on Mental Illness. (2017a). Mental health conditions. http://www.nami.org/Learn-More/Mental-Health-Conditions
- National Alliance on Mental Illness. (2017b). Mental illness facts. https://namica.org/resources/mental-illness/mental-illness-facts/
- National Alliance on Mental Illness. (2017c). Securing stable housing. http://www.nami.org/Find-Support/Living-with-a-Mental-Health-Condition/Securing-Stable-Housing
- Newhouse, J. P., & Insurance Experiment Group. (1993). Free for all? Lessons from the RAND Health Insurance Experiment. Cambridge, MA: Harvard University Press.
- O’Day, B., Blyler, C., Fischer, B., Gill, C., Honeycutt, T., Kleinman, R., . . . Wishon-Siegwarth, A. (2014). Improving employment outcomes for people with psychiatric disorders and other disabilities. https://aspe.hhs.gov/execsum/improving-employment-outcomes-people-psychiatric-disorders-and-other-disabilities
- Organisation for Economic Co-operation and Development. (2012). Sick on the job? Myths and realities about mental health and work. Paris, France: OECD.
- Organisation for Economic Co-operation and Development. (2014). Mental health and work: United Kingdom. Paris, France: OECD.
- Organisation for Economic Co-operation and Development. (2015). The OECD mental health and work policy framework. Paris, France: OECD.
- Padgett, D. K., Patrick, C., Burns, B. J., Schlesinger, H. J., & Cohen, J. (1993). The effect of insurance benefit changes on use of child and adolescent outpatient mental health services. Med Care, 31(2), 96–110.
- Perneger, T. V., Allaz, A. F., Etter, J. F., & Rougemont, A. (1995). Mental health and choice between managed care and indemnity health insurance. American Journal of Psychiatry, 152(7), 1020–1025.
- Reed, L. S. (1974). Utilization of care for mental disorders under the Blue Cross and Blue Shield plan for federal employees, 1972. American Journal of Psychiatry, 131(9), 964–975.
- Robert Wood Johnson Foundation. (2011). Mental disorders and medical comorbidity. New York, NY: Robert Wood Johnson Foundation.
- Rogers, W. H., Wells, K. B., Meredith, L. S., Sturm, R., & Burnam, M. A. (1993). Outcomes for adult outpatients with depression under prepaid or fee-for-service financing. Archives of General Psychiatry, 50(7), 517–525.
- Rosenheck, R., Leslie, D., Keefe, R., McEvoy, J., Swartz, M., Perkins, D., . . . CATIE Study Investigators Group. (2006). Barriers to employment for people with schizophrenia. American Journal of Psychiatry, 163(3), 411–417.
- Social Security Administration. (2017a). Employment networks in Social Security’s Ticket to Work program. https://www.ssa.gov/pubs/EN-05-10065.pdf
- Social Security Administration. (2017b). Working while disabled: How we can help. https://www.ssa.gov/pubs/EN-05-10095.pdf
- Social Security Administration, Office of the Chief Actuary. (2017). Selected data from Social Security’s disability program. https://www.ssa.gov/oact/STATS/dibStat.html
- Substance Abuse and Mental Health Services Administration. (2014). Behavioral health spending and use accounts, 1986–2014. Rockville, MD: U.S. Department of Health and Human Services.
- Substance Abuse and Mental Health Services Administration. (2015). Homelessness and housing Retrieved from https://webcache.googleusercontent.com/search?q=cache:DnWLobTaCbIJ:https://www.samhsa.gov/homelessness-housing+&cd=1&hl=en&ct=clnk&gl=us.
- Taube, C. A., Kessler, L. G., & Burns, B. J. (1986). Estimating the probability and level of ambulatory mental health services use. Health Services Research, 21(2), 321–340.
- Tinelli, M., & Kanavos, P. (2015). Cost and impact of non-treating severe mental illnesses (SMIs): The case study of schizophrenia. http://www.braincouncil.eu/wp-content/uploads/2015/07/LSE-publication-Impact_Not_Treating_Schizophrenia_EBC17June2015.pdf
- U.S. Department of Health and Human Services, Agency for Healthcare Research and Quality. (2015). Economics and improvements in behavioral health. Population Health: Behavioral and Social Science Insights Health.
- U.S. Department of Health and Human Services, National Institute of Mental Health. (2015). Any mental illness (AMI) among U.S. adults.
- U.S. Department of Health & Human Services, Office of the Assistant Secretary for Planning and Evaluation. (2014). Medicaid and permanent supportive housing for chronically homeless individuals: Emerging practices from the field: 5.5. Payment structures and administrative challenges.
- U.S. Department of Health and Human Services, Substance Abuse and Mental Health Services Administration. (2016). Serious mental illness among adults below the poverty line. National Surveys on Drug Use and Health (NSDUH).
- U.S. Department of Housing and Urban Development. (2017). Section 811 Supportive Housing for Persons with Disabilities Program.
- Umberson, D., & Montez, J. K. (2010). Social relationships and health: A flashpoint for health policy. Journal of Health and Social Behavior, 51 Suppl, S54–66.
- Wang, P. S., Simon, G. E., Avorn, J., Azocar, F., Ludman, E. J., McCulloch, J., et al. (2007). Telephone screening, outreach, and care management for depressed workers and impact on clinical and work productivity outcomes: A randomized controlled trial. JAMA, 298(12), 1401–1411.
- Watts, C. A., Scheffler, R. M., & Jewell, N. P. (1986). Demand for outpatient mental health services in a heavily insured population: The case of the Blue Cross and Blue Shield Association’s Federal Employees Health Benefits Program. Health Services Research, 21(2 Pt 2), 267–289.
- Webber, M., Corker, E., Hamilton, S., Weeks, C., Pinfold, V., Rose, D., . . . Henderson, C. (2014). Discrimination against people with severe mental illness and their access to social capital: Findings from the Viewpoint survey. Epidemiology and Psychiatric Sciences, 23(2), 155–165.
1. Parity in the US context means that coverage for mental and substance use disorders can be no more restrictive than that for medical-surgical conditions. These provisions were set out in the Mental Health Parity and Addictions Equity Act of 2008.