Racial-Ethnic Disparities in Outpatient Mental Health Care in the United States

In the United States, studies have shown that non-Hispanic Black and Hispanic individuals are less likely than non-Hispanic White individuals to receive outpatient mental health care (1, 2). Key drivers of these differences are thought to operate through systemic and social factors, including structural racism and discrimination (3) at individual, community, and population levels (4). One pathway through which structural racism might adversely affect access to mental health care is through diminished opportunities for financial advancement and lower availability of resources, as reflected in education, employment, income, and health insurance (5). Yet, little is known about recent national racial-ethnic patterns in outpatient mental health care and their relationship to socioeconomic status (6). Although reducing racial-ethnic differences in mental health service access would decrease downstream disparities in mental health care delivery, after entering treatment, Black and Hispanic patients may also receive less intensive or less effective mental health care than White patients (7, 8).

The current evidence concerning differences in minimally adequate mental health care shows stronger and more consistent disparities between Blacks and Whites than between Hispanics and Whites (8–10). The Health Care for Communities Survey found that Black patients with probable depressive or anxiety disorders were less than half as likely as White patients to receive minimally adequate mental health care (10). The Midlife Development in the United States Survey (8) and the National Comorbidity Survey–Replication (9) yielded similar results. More recently, commercially insured Black patients who received psychotherapy were reported to receive fewer sessions than their White counterparts (11).

In this study, we aimed to update national population-based estimates of outpatient mental health care for Black, Hispanic, and White people in the United States and reassess their relationship to socioeconomic status. Among individuals receiving mental health treatment, we also report racial-ethnic differences in patterns and adequacy of care. A greater understanding of these two aspects of mental health care might inform efforts to address drivers of racial-ethnic disparities in care. We anticipated that Hispanic and Black individuals would be less likely than White individuals to receive outpatient mental health care and, among those receiving care, that Hispanic and Black patients would be less likely than White patients to receive minimally adequate care.

Methods

Data Sources

Data were analyzed from the 2018–2019 Medical Expenditure Panel Survey (MEPS) of U.S. households conducted by the Agency for Healthcare Research and Quality. Technical information concerning the fielding and nonresponse adjustment of the MEPS is provided in previous studies (12, 13). The MEPS was administered in English and Spanish by computer-assisted personal interviewing. A new nationally representative sample of households is selected each year, and participants are interviewed five times over a 2-year period to improve the recall of service use. The MEPS had full-year response rates of 42.7% in 2018 and 39.5% in 2019, with >90% of surveys administered in person, and oversampled Black and Hispanic individuals.

Study Samples

The first analytic sample included all individuals in the household survey sample, ages ≥4 years, from the three racial-ethnic groups of interest to this study (N=49,731). Within this sample, we examined racial-ethnic differences in rates of any outpatient mental health care use. The second sample was limited to people who used outpatient mental health care. This sample was examined for racial-ethnic differences in the type of outpatient mental health care used and whether individuals received minimally adequate care.

Mental Health Service Outcomes

Mental health service outcomes included any outpatient mental health care and, among people with outpatient mental health care, included receipt of psychotropic medication, psychotherapy, or both and receipt of minimally adequate treatment. Outpatient mental health care was defined as having at least one visit with a psychiatric diagnosis (ICD-10-CM codes F10–F69 and F90–F99); at least one visit to a psychiatrist, social worker, psychologist, or mental health counselor; or prescription of at least one psychotropic medication for a mental health condition. Because psychological distress is commonly diagnosed outside of specialty mental health settings (14–16) where F codes may not be used, outpatient visits with other selected ICD-10-CM codes (i.e., for insomnia, substance use counseling, life management difficulties, social skill inadequacy, attention and concentration deficits, stress, fatigue, and postpartum mood disturbance) were included as evidence of outpatient mental health care (see Table S1 in the online supplement to this article).

Psychotherapy was defined as “a treatment technique for certain forms of mental disorders relying principally on talk/conversation between the mental health professional and the patient.” It specifically included “individual, family, and/or group therapies” (17). Psychotropic medications included antidepressants, antipsychotics, anxiolytics and hypnotics, stimulants or other medications for attention-deficit hyperactivity disorder (ADHD), and mood stabilizers obtained from pharmacies during the survey year (see Table S2 in the online supplement). To facilitate comparison with previous research, we defined minimally adequate care as having at least eight visits to a mental health provider during the study year or having at least four visits to any provider and receiving prescribed antidepressants (for depression and anxiety), an anxiolytic or hypnotic medication (anxiety), or a stimulant or other ADHD medication (ADHD or disruptive behavior disorder) (8–10).

Mental Health Conditions and Other Characteristics

The MEPS collected information on psychiatric conditions for each outpatient visit, and professional coders assigned ICD-10-CM categories (18). Mental health conditions included mood disorders with a depression subgroup, anxiety disorders, trauma- and stress-related disorders, ADHD and other disruptive behavior disorders, other mental disorders, and mental health–related conditions (see Table S1 in the online supplement). Other characteristics included age in years, sex, marital status, educational attainment, family income, health insurance, employment status, and perceived mental health status (excellent, very good, good, fair, or poor).

Statistical Analysis

A preliminary analysis compared the background characteristics of White, Black, and Hispanic persons ages ≥4 years (see Table S3 in the online supplement). This analysis was followed by a comparison, by racial-ethnic group, of annual rates of any outpatient mental health care overall and stratified by age group and sex. Differences in rates and associated 95% CIs were calculated and adjusted for perceived mental health status. In these analyses, we withheld adjustment for other sociodemographic characteristics, such as income or education, which may mediate the association between race-ethnicity and the mental health care outcomes (19–21). Stepped regressions were fit with race-ethnicity as the independent variable of interest and any outpatient mental health care as the dependent variable. After unadjusted regressions, age, sex, and perceived mental health status were stepped in, followed by education, income, health insurance, and employment. Among individuals receiving outpatient mental health care (hereafter “patients”), the racial-ethnic groups were compared with respect to sociodemographic characteristics and clinical background. Next, the racial-ethnic patient groups were compared with respect to the proportions of individuals receiving psychotherapy, psychotropic medication, or both, controlling for age, sex, and perceived mental health status by using adjusted mean differences. Finally, the racial-ethnic patient groups treated for depression, anxiety, and ADHD or disruptive behavior disorders were compared with respect to receiving selected psychotropic medications and adequate mental health treatment. In sensitivity analyses, these regressions were also controlled for health care coverage. In evaluating the results, it is important to bear in mind that race and ethnicity are imprecise social constructs without biological meaning (22).

All statistical analyses were performed in R, version 4.1.2, by using the survey package, version 4.0, to accommodate the complex sample design and weighting of observations. The institutional review board of the New York State Psychiatric Institute exempted this analysis from human subjects review.

Results

Annual Rates of Outpatient Mental Health Care Utilization

In the overall national population (White, N=29,126; Black, N=7,965; and Hispanic, N=12,640 [Table 1]), fair or poor perceived mental health was significantly less common among Hispanic (6.7%) than White (7.9%) individuals but similar to Black (7.5%) individuals (percentages are weighted percentages; see Table S3 in the online supplement). The annual unadjusted rate of outpatient mental health care utilization was more than twice as high for White (25.3 per 100 persons) as for Black (12.2 per 100 persons) or Hispanic (11.4 per 100 persons) individuals (Table 1). In analyses adjusted for perceived mental health status, White males and females in each age group had significantly higher rates of outpatient mental health care compared with their Black counterparts (Table 1). Similarly, compared with Hispanic people, White individuals had significantly higher rates of outpatient mental health care. For all ages and among individuals ages 18–44 and 45–64 years, Black men had higher rates of outpatient mental health care than Hispanic men, but the reverse was true for men ages ≥65.

Socioeconomic Mediation of Racial-Ethnic Differences in Outpatient Care

In the stepped regression analyses, differences in outpatient mental health care utilization between Black and White individuals and between Hispanic and White people changed little after we controlled for age, sex, perceived mental health status, education level, family income, health insurance, and employment status (Table 2).

Characteristics of People Using Outpatient Mental Health Care

In unadjusted analyses, several sociodemographic and clinical differences emerged among White, Black, and Hispanic individuals who used outpatient mental health care (Table 3). Compared with White patients, Black and Hispanic patients were significantly more likely to be ages 4–17 and single. They were also more likely to have a low family income, public insurance (for ages <65), or fair or poor perceived mental health and less likely to have a high school diploma or equivalent or to be employed (for ages <65). Compared with White patients, Black and Hispanic patients were significantly less likely to be treated for anxiety and depression, whereas Black patients were more likely than White or Hispanic patients to be treated for trauma- and stress-related disorders.

Psychotropic Medications and Psychotherapy

We observed several racial-ethnic differences in the use of psychotropic medications or psychotherapy (Table 4). After we controlled for age, sex, and perceived mental health status, we found that Black and Hispanic patients were significantly less likely than White patients to receive medications but were more likely to receive psychotherapy (Table 4). Black patients were also significantly more likely than Hispanic patients to receive psychotherapy.

In line with this overall pattern, Black patients who were treated for depression or other mood disorders, anxiety, trauma- or stress-related disorders, or mental health–related conditions were less likely than White patients to receive psychotropic medications. However, Black patients treated for depression or other mood disorders, anxiety, or ADHD or disruptive behavior disorders were significantly more likely than White patients to receive psychotherapy. Black patients treated for depression or other mood disorders were also significantly more likely than White patients to receive both psychotherapy and psychotropic medications.

Compared with White patients, Hispanic patients treated for mood, anxiety, or trauma- or stress-related disorders; ADHD or disruptive behavior disorders; or mental health–related conditions were less likely to receive a psychotropic medication (Table 4). Hispanic patients treated for ADHD or disruptive behavior disorders were significantly more likely than White patients to receive psychotherapy. Finally, Black patients who were treated for depression were more likely than Hispanic patients to receive psychotherapy.

Psychotropic Medications and Minimally Adequate Treatment

After we controlled the analysis for age, sex, and perceived mental health status, among patients treated for depression, White patients were significantly more likely than Black or Hispanic patients to receive antidepressants (Table 5). These differences persisted after we also controlled for health care coverage (see Table S4 in the online supplement). Compared with White patients treated for depression, Black patients were more likely to receive antipsychotic medications, and this difference persisted after we controlled for health care coverage. No significant racial-ethnic differences were found in the percentages of patients who received minimally adequate treatment.

Among patients treated for anxiety disorders, White patients were more likely than Black or Hispanic patients to receive antidepressants, and these differences endured after adjustment for perceived mental health status and health care coverage (Table 5). However, White patients who were treated for anxiety were less likely than Black patients to receive antipsychotic medications, but not after we also controlled for health care coverage. White patients treated for anxiety disorders were less likely than Hispanic patients to receive anxiolytics or sedatives, but not after we also controlled for health care coverage (see Table S4 in the online supplement). White patients treated for ADHD or disruptive behavior disorders were significantly more likely than Hispanic patients to receive stimulants or other ADHD medications, including after we controlled for health care coverage (see Table S4 in the online supplement).

Discussion

Our findings reveal that in the United States, Black and Hispanic individuals were less than half as likely as White individuals to receive any outpatient mental health care service in a given year. These gaps in service use, which are broadly consistent with observations from earlier epidemiological surveys (1, 2), were evident across different age groups and both sexes, persisted after we controlled for several potentially confounding socioeconomic factors, and highlight the need for renewed efforts to improve mental health service access for Black and Hispanic people. Beyond marked racial-ethnic differences in access to care, Hispanic, Black, and White people also differed in their likelihood of receiving psychotherapy and psychotropic medications. Contrary to our expectations, however, for several common mental disorders, these groups did not significantly differ in their likelihood of receiving minimally adequate treatment once treatment was initiated. However, persisting racial-ethnic gaps in access to mental health care underscore the need for service reforms to facilitate greater access to timely, appropriate, and high-quality mental health care that is tailored to the culture and context of each patient.

As a result of unjust social and economic policies and practices (3), substantial socioeconomic differences exist among Black, Hispanic, and White individuals who receive outpatient mental health care in the United States. Relative to White patients, Black and Hispanic patients tended to be younger and to have less education, have lower incomes, and lack employment and were more likely to have public health insurance. Because many mental health specialists (23), including most psychiatrists (24), do not accept patients with Medicaid, socioeconomic disadvantages disproportionately restrict the mental health care options available to Black and Hispanic patients.

Among patients receiving treatment for depression, anxiety, or ADHD or disruptive behavior disorders, no significant racial-ethnic differences were found in the likelihood of receiving minimally adequate treatment. Previous reports of Black and Hispanic individuals having lower likelihoods than White individuals of receiving minimally adequate mental health treatment (8–10) were likely driven primarily by racial-ethnic differences in access to any outpatient mental health care and only to a lesser extent by differences in service patterns after treatment initiation. These patterns underscore the central importance of maintaining a focus on expanding service accessibility to underserved racial-ethnic groups. Yet, low rates of receiving even minimally adequate care across racial-ethnic groups also raise broader concerns about general deficiencies in the quality and effectiveness of community mental health care. Consistent with pervasive suboptimal care, one nationally representative study recently reported that only 45% of White, 32% of Black, and 28% of Hispanic U.S. adults who were receiving antidepressants for depression were in symptomatic remission (25).

Among individuals receiving outpatient mental health care, Black and Hispanic patients tended to be less likely than White patients to receive psychotropic medications and more likely to receive psychotherapy. Beyond a general population-wide preference for psychotherapy over pharmacological treatments (26), Black people, and to a lesser extent Hispanic people, prefer psychotherapy over medications for the treatment of depression more often than do White people (27–29). These differences may be partially mediated by racial-ethnic differences in attitudes concerning the etiology of depression, the safety and effectiveness of antidepressants and psychotherapy, and the role of prayer and spirituality in coping with an episode of depression (23). Because accommodating patient preferences is related to improved treatment adherence and outcomes (30), the broad alignment of racial-ethnic preferences with observed treatment patterns may capture a degree of responsiveness of clinical services to patients’ preferences. Adapting mental health care to patients’ cultural values and contexts has also been found to increase use and improve outcomes (31).

Among patients treated for depression or anxiety, Black patients were more likely than White patients to receive antipsychotic medications. In line with this finding, Black patients with Medicaid who initiated treatment for major depressive disorder have also been found to be significantly more likely than White patients with Medicaid to receive antipsychotic medications (32). The metabolic risks of antipsychotic medications and their small-to-modest effects as adjunctive treatments for major depressive disorder (33) have prompted concerns regarding overuse of these medications (34). Higher rates of obesity and diabetes among Black patients, compared with White patients (35, 36), compound these safety concerns for Black patients and highlight the need to probe the basis of this racial-ethnic variation in pharmacological treatment. Specifically, it is important to evaluate the extent to which differences between Black and White patients in antipsychotic treatment for nonpsychotic disorders represent an appropriate response to variation in illness severity or comorbid diagnoses (32) or reflect biased clinical overvaluation of psychotic symptoms among Black patients with mood disorders (37). This currently unexplained difference is a potential target for quality improvement efforts.

In treatment for ADHD or disruptive behavior disorders, Hispanic patients were significantly less likely than White patients to receive stimulants or other ADHD medications. Previous research has documented similar patterns for children (38). Mental health professionals may be able to better serve Hispanic parents and children by developing partnerships with schools, churches, and traditional healers from whom these individuals may initially seek care (39) and by implementing culturally sensitive mental health interventions (40). Consistent with an analysis of Medicaid claims (41), Black and Hispanic patients with ADHD were more likely than White patients to receive psychotherapy. Because psychotherapy tends to yield smaller effects than stimulants in reducing ADHD symptoms (42), it is important that all patients and their families be provided with relevant clinical information to make informed choices about available ADHD treatments.

Our analyses had several limitations. First, the MEPS relies on respondent recall and diaries that may underestimate mental health care in a manner that varies across racial-ethnic groups. Second, diagnoses in MEPS are based on respondent reports and are not subject to expert validation. Third, the perceived mental health status measure is only modestly correlated with symptoms of psychological distress (43), and this correlation may vary by racial-ethnic group (44) because of differences in mental health literacy, cultural attitudes, stigma, or other factors. Fourth, children and adults, who were combined into a single group to ensure sufficient statistical power, vary in their mental health treatment patterns. Fifth, the analyses did not capture important regional variations in policies, resources, or social conditions that play a central role in health disparities (45). Sixth, left and right censoring related to the survey year truncated some treatment episodes. Finally, the MEPS provides no information on clinical outcomes, and it is therefore not possible to evaluate associations among race-ethnicity, the adequacy of care, and mental health outcomes.

Conclusions

Large national racial-ethnic differences in use of mental health services underscore persistent disparities in service access. Structural racism (46), cultural attitudes, and racial-ethnic biases in clinical recognition of mental health problems likely contribute to low rates of mental health treatment among Black and Hispanic individuals. Although significant racial-ethnic differences were also evident in the treatments received after patients entered mental health treatment, meaningful progress in narrowing racial-ethnic differences in mental health service delivery will require dedicated efforts to promote service access for underserved minority groups. These efforts may necessitate developing culturally centered integrated primary care and specialty mental health services that help nurture trusting patient-provider partnerships in environments that respect and appreciate patients’ ethnic and cultural diversity. Advocacy may also be required to address deep structural factors, such as the shortage of psychiatrists in areas that have high proportions of Black and Hispanic individuals (47), which may perpetuate racial-ethnic inequities in access to mental health care.