Racial disparities in mental health outcomes have been widely documented in noninstitutionalized community psychiatric samples, but few studies have specifically examined the effects of race among individuals with the most severe mental illnesses. A sample of 925 individuals hospitalized for severe mental illness was followed for a year after hospital discharge to examine the presence of disparities in mental health outcomes between African American and white individuals diagnosed with a severe psychiatric condition. Results from a series of individual growth curve models indicated that African American individuals with severe mental illness experienced significantly less improvement in global functioning, activation, and anergia symptoms and were less likely to return to work in the year following hospitalization. Racial disparities persisted after adjustment for sociodemographic and diagnostic confounders and were largely consistent across gender, socioeconomic status, and psychiatric diagnosis. Implications for social work research and practice with minorities with severe mental illness are discussed.
KEY WORDS: ethnicity; mental health outcomes; race; schizophrenia; severe mental illness
Race has been described as one of the most critical factors for understanding social problems in the contemporary United States (Gallagher, 2007). Within the past several decades, social work researchers have begun turning to the issue of race to understand the disparities related to mental health care of those who are the most disabled in our society (Newhill & Hams, 2008). In a comprehensive report on the status of mental health care for racial and ethnic minorities in the United States, former Surgeon General David Satcher concluded that minorities suffer a disproportionate burden of mental illness because they often have less access to services than other Americans, receive lower quality care, and are less likely to seek help when they are in distress. The report concluded that there is a large gap between the need for services and the services actually provided. Major contributors to these disparities include poverty, stigma, and discrimination (Rank, 2004; U.S. Department of Health and Human Services [HHS], 2001; Wilson, 2009).
Although the documentation of racial disparities in mental health services and outcomes among the general population of people with mental illness provided by the surgeon general's report was an important advance for the field, little is known about how these broader findings apply to individuals with severe and persistent mental illness. Severe and persistent mental illnesses (for example, schizophrenia, bipolar disorder, and recurrent major depression) often produce the most serious and long-term psychosocial disabilities (Murray & Lopez, 1996). Many individuals with these illnesses also live in poverty and thus are usually dependent on public mental health services for care (Hudson, 2005). The question of whether race remains a factor in this disabled and frequently impoverished population is of critical importance for ensuring that effective services are available to those most in need. To date, however, most research on racial disparities in mental health services and outcomes has focused on noninstitutionalized community samples, such as the National Comorbidity Studies (Kessler, Mcgonagle, Zhao, & Nelson, 1994) and the Healthcare for Communities Survey (Sturm et al., 1999), which frequently excluded those with the most severe and persistent mental illnesses, such as schizophrenia.
Of the studies that have examined racial disparities and severe mental illness, most have focused almost exclusively on the overdiagnosis of certain mental health conditions and disparities in the use of certain forms of treatment, such as inpatient care and depot pharmacotherapy. These studies have consistently found that African Americans with severe mental illness tend to be disproportionately diagnosed with psychotic conditions such as schizophrenia, even though there is little to no difference in actual prevalence rates (Buchanan & Carpenter, 2005), and less frequently diagnosed with mood disorders compared with their white counterparts (Barnes, 2008; Snowden & Cheung, 1990; Strakowski, Shelton, & Kolbrener, 1993). In addition, racial minorities, particularly African Americans, are prescribed more psychotropic medications at higher doses, are more likely to receive injectable medicines (Citrome, Levine, & Ailingham, 1996; Segal, Bola, & Watson, 1996), and are more likely to be hospitalized involuntarily than white Americans (Rosenfield, 1984). Despite these differences, some studies have shown that racial minority individuals with severe mental illness use similar (if not higher) rates of inpatient and outpatient mental health services (Folsom et al., 2007; Kilboume et al., 2005), although such individuals are disproportionately underrepresented in case management services (Barrio et al., 2003). In general, research has shown that at some of the lowest sodoeconomic levels, the service gap between white and nonwhite people can shrink (for example, Alegrla et al., 2002). One compelling explanation for this phenomenon is that most people with severe mental illness are eligible for Medicaid, which may equalize access to care by removing the well-documented racial disparities in private insurance coverage (Snowden & Thomas, 2000).
Although the picture of racial disparities in mental health services among people with severe mental illness is clearly complex and evidence is only recently emerging with regard to Hispanic and Latino populations, findings have consistently indicated that African Americans with severe mental illness receive a different standard of care. The treatment of such individuals disproportionately consists of greater use of inpatient psychiatric services (Rosenfield, 1984; Snowden, Hastings, & Alvidrez, 2009), injectable medications (Shi et al., 2007), and less costly outpatient treatment options (Kuno & Rothbard, 2002). Although this does not necessarily mean that African Americans receive less care, the care they do receive may be suboptimal and less able to meet their needs. Unfortunately, the critical question of whether African Americans with severe mental illness experience less favorable mental health treatment outcomes has been infrequently addressed. One study by Gift, Harder, Ritzler, and Kokes (1985) examined 217 individuals hospitalized for a severe mental illness and followed these individuals for two years post--hospital discharge. They found that at two years' follow-up African American participants had greater psychotic symptoms after discharge than white participants. However, no information was provided on how these symptoms, which were significantly elevated at baseline, might have differentially changed over time between races. Another study by Chinman, Rosenheck, and Lam (2000) found that in a large sample of individuals with severe mental illness, white participants receiving case management services tended to experience greater reduction in symptoms than African Americans. Although such studies point to the possibility of disparities in important mental health outcomes between African American and white individuals with severe mental illness, most have made use of cross-sectional designs and have generally used modest sample sizes. In addition, few studies have examined outcomes other than symptom reduction, such as level and quality of community psychosocial functioning and employment, and most have not accounted for the confounding effects of socioeconomic status (SES) or differential psychiatric diagnosis.
In this research, we used individual growth curve modeling in a large longitudinal sample (N = 925) of African American and white individuals with severe mental illness to examine the presence of racial disparities in mental health outcomes after psychiatric hospital discharge. The study participants were then assessed in the community every 10 weeks for a year on a number of broad symptom and functioning domains, and racial differences in rates of improvement in these domains were examined after adjusting for socioeconomic, diagnostic, and other potential confounding variables.
This research was conducted using existing data collected as part of the MacArthur Violence Risk Assessment Study, the methods of which have been described in detail elsewhere (Monahan et al., 2001). Participants included individuals recruited from psychiatric inpatient units in three cities (Pittsburgh, Pennsylvania; Kansas City, Missouri; and Worcester, Massachusetts). Participants were included in the MacArthur Violence Risk Assessment Study if they had been hospitalized for less than 21 days, spoke English as a primary language, were between the ages of 18 and 40, and carried a medical chart diagnosis of schizophrenia, schizophreniform disorder, schizoaffective disorder, major depression, dysthymia, bipolar disorder, brief reactive psychosis, delusional disorder, alcohol or other drug abuse or dependence, or a personality disorder. In total, 1,695 patients were recruited, 1,136 (71%) of whom agreed to participate. Individuals who agreed to participate were younger, less likely to be diagnosed with schizophrenia, and more likely to be diagnosed with a substance use or personality disorder (Steadman et al., 1998). Of those individuals who participated, 925 (81%) had complete data available for at least one follow-up period and were either African American or Caucasian (only 2% [n = 21] of the sample was Hispanic, and thus these individuals were excluded from further analysis).
Among the 925 participants included from the MacArthur Violence Risk Assessment Study, 70% were Caucasian and 30% were African American. Ages ranged between 18 and 40 years (M= 29.90, SD= 6.20), and 58% of the participants were male. Psychiatric diagnosis was...