Racial and Ethnic Differences in Etiology/Risk Factors

Structural inequities that disadvantage families of color are numerous. Examples include housing policies (e.g., redlining, home loan financing), drug and criminal justice policies (e.g., treatment of crack versus powder cocaine), employment policies (e.g., exclusion of agricultural and domestic workers from unemployment and retirement benefits) and disinvestment in communities with a high proportion of Black, Hispanic, and Native American residents.²⁰¹ In addition to separating children from their parents, mass incarceration due to drug and criminal justice policies directly affect families’ mental health by increasing the risk of poverty and adverse childhood experiences.²⁰² These factors have all contributed to wealth inequity wherein White Americans had almost 10 times the net worth of Black Americans²⁰³ in 2016. These policies are also reflected in the healthcare system; a recent study examining capital assets of 4,476 Medicare-participating found that hospitals serving people of color had lower capital assets (-$215,121/bed, P < .0001) and recent purchases (-$83,608/bed, P < .0001). They were also less likely to offer 19 of 27 specific capital-intensive services,²⁰⁴ based on data from 2013–2017.

The challenges posed by these and other structural challenges, as well as the resulting income inequality, has a damaging impact on mental health in disadvantaged communities. Income inequality is correlated with depression prevalence,^{205, 206} risk of death from drug overdose,²⁰⁷ and child well-being.²⁰⁸ The estimated number of deaths attributable to six social factors (low education, racial segregation, low social support, individual-level poverty, income inequality, area-level poverty) in the United States is comparable to the number of deaths attributed to pathophysiological and behavioral causes.²⁰⁹ While increasing awareness has led to commitments by many organizations and municipalities to mitigate inequities, the recent COVID-19 pandemic is likely to exacerbate existing inequities,²¹⁰ at least in the short term.

Although some research indicates that a strong sense of ethnic identity and job security may be protective factors among Black and Latinx Americans,²¹¹ the interpersonal experience of racism and discrimination has a deleterious impact on mental as well as physical health. In fact, an overview of reviews found that association between discrimination and mental health to be stronger than the association with physical health.²¹² In this review, discrimination was associated with higher levels of depression, anxiety symptoms, and psychological distress as well as with meeting criteria for mental health disorders. This review also found that discrimination increased the risk of health behaviors that impact mental health, including unhealthy use of alcohol and sleep disturbance.

Further, persons reporting experiences of racial discrimination were less trusting of health care workers and systems, had lower satisfaction with care, had lower ratings of patient-provider communication and relationships, lower adherence to medical recommendations, and delays in seeking health care. For example, 32% of Black people, 20% of Latinx people, and 23% of Indigenous people report avoiding medical care because of experiences of personal discrimination due to their race or ethnicity in health care settings.²¹³ Similarly, a large population-based survey of older California residents found that Latinx and Asian or Pacific Islander respondents were more likely to endorse feeling uncomfortable talking to clinicians as a reason for not seeking treatment for mental health issues than White respondents.²¹⁴

Racial and Ethnic Differences in Prevalence and Burden of Mental Health Diagnoses

According to 2008-2012 NSDUH data, multiracial adults had the highest prevalence of having a mental health condition (24.9%), followed by Native American/Alaska Native (22.7%), White (19.0%), Black (16.8%), Latinx (15.3%), and Asian/Asian-American (13.4%) adults. However, some have noted that presentation of depression can differ between Black and White Americans.²¹¹ For example, studies have identified differences in expression of somatic and sleep symptoms, physical functioning, and psychosocial distress,²¹⁵ differences in core symptom presentation (sadness/depression vs. irritability/agitation), and differences in verbal descriptions of their mood between Black and White Americans. Thus, slightly lower prevalence could reflect some level of underdiagnosis among Black Americans. Cross-sectional NSDUH data from 2015-2019 showed that although Black, Latino, and Asian adults had lower lifetime and past-year rates of major depressive episodes (MDE) than White adults (lifetime 8.9 %, 10.2%, and 8.3% vs. 15.6%; past-year 5.4%, 5.7%, and 8.3 vs. 7.8%), Black and Latino adults had higher rates of persistence (experiencing lifetime and past-year MDE) and severity (experiencing past-year MDE with very severe impairment) compared with White adults (persistence 60.5% and 56.2% vs. 50.1%; severity 23.9% and 19.4% vs. 18.9%).²¹⁶ Thus, commonly used interview tools may not fully capture the burden of depressive disorders equally among different race and ethnic groups.

Racial and Ethnic Differences in Rates of Screening for Mental Health Conditions

We found very limited information on real-world mental health screening rates in different racial and ethnic groups. One study within an urban New England healthcare system found that, over the years 2010-2012, Latinx patients were more likely to be screened for depression (60-66% were screened) than White patients (55-59% screened), who were in turn more likely to be screened than patients who were Black (53% screened) or of Asian descent²¹⁷ (52-55% screened). A smaller but more recent study examined screening rates in a nationally representative group of 1,852 patients who had completed both the Preventive Care Self-Administered Questionnaire (PSAQ) and the MEPS Self-Administered Questionnaire in 2014-2015.²¹⁸ This study found that the percent who were assessed for depression varied by race and ethnicity, with lower rates among Hispanic (31.3% assessed), Non-Hispanic Black (38.1%), and Asian American (27.7%) respondents than among Non-Hispanic White respondents (54.4%, p<.05 for all males and females in all 3 race and ethnic groups).

What Is Known About the Validity of the Most Commonly Used or Recommended Instruments to Screen for Depression, Anxiety, and Suicide Risk in U.S. Racial/Ethnic Minority Patients? (Contextual Question 5)

One potential factor contributing to misdiagnosis could be the screening instruments themselves. Of particular interest for screening instruments is whether they are equally valid for all racial and ethnic groups in the United States, and therefore appropriate for population-based screening. A recent publication²¹⁹ examined the factor structure and measurement invariance of the PHQ-9 using a representative sample of over 30,000 US adults. The authors organized participants into six major groups: non-Hispanic White, non-Hispanic Black, Mexican American, other Hispanic, non-Hispanic Asian, and other/multi-racial. A two-factor structure was identified, with a cognitive/affective factor and a somatic factor. The authors further determined that the factors had the same meaning for the six major racial/ethnic groups and using the PHQ-9 to compare depression symptoms among these racial/ethnic groups was appropriate. An additional study compared the validity of the PHQ-9 among American Indian and Caucasian American patients.²²⁰ They also determined a two-factor model was more appropriate than a single factor, and concluded it was acceptable to compare PHQ-9 scores among Caucasian American and American Indian population groups.

A smaller, older study also examined the validity of the PHQ-9 among different racial and ethnic primary care patients (n=5,053): non-Hispanic White, African American, Chinese American, and Latino.²²¹ They identified a single factor structure and similar mean scores for the four racial/ethnic groups. There were some minor differences between groups in the pattern of symptom endorsement. Relative to Non-Hispanic White Americans, Chinese Americans were more likely to endorse sleep changes and psychomotor symptoms and less likely to have appetite concerns. Latino participants were relatively more likely to endorse anhedonia and less likely to endorse sleep changes, appetite changes, and guilt. Item loading was very similar between Non-Hispanic White and Black participants. Despite the minor differences in symptom endorsement, the authors concluded that among the four racial/ethnic groups, the PHQ-9 measured a common concept of depression.

With respect to the other depression screening instruments, a meta-analysis of 28 studies determined that the original four-factor structure of the CES-D was not the best fit for all five racial/ethnic groups examined: African American, American Indian, Asian, White, and Hispanic.²²² This suggests that the CES-D is not measuring the same symptoms among these racial/ethnic groups. In a study looking at predictors of future MDD diagnosis, the presence of depressive symptoms based on the CES-D was a strong predictor for White participants.²²³ For Black participants, however, a single-item self-rated mental health measure was a strong predictor of future MDD while CES-D score was not associated with future MDD. These associations controlled for demographic and socioeconomic variables and suggest that the CES-D may not be as accurate at capturing important aspects of depression in Black participants.

One of our included studies⁴⁸ (n=492) examined the impact of race on the measurement properties of the GDS-15 and found no difference between racial/ethnic groups (non-Hispanic White, non-Hispanic Black, and Hispanic). However, a meta-analysis²²⁴ with 26 studies determined that the factor structure differed when it was administered in languages other than English (Chinese, Greek, Hindi, Iranian, Italian, Japanese, Korean, Portuguese, and Turkish). A 2017 study²²⁵ (n=2,687) also determined that while race/ethnicity (Black, Hispanic-Latina, Asian, White) did not impact the psychometric properties of the EPDS, socioeconomic status did and suggested that differences others have identified between ethnic and racial groups might be due to socioeconomic status disparities.

A study²²⁶ from 2015 examined the cultural biases of the GAD-7 among 950 participants in the US and Canada. Participants were categorized into four racial/ethnic groups: White Americans, Hispanic, Black/African American, and White Canadians. The total score for the GAD-7 was significantly different for Black/African Americans when compared to the three other groups, but no differences were seen between the other racial and ethnic groups. Further, the original single factor structure was only a good fit for the White American group; when covariance was allowed for three items, the fit improved for all racial/ethnic groups. Further, individuals identifying as Black/African American with high GAD symptoms did not endorse three items as strongly as other racial and ethnic participants with a similar level of GAD symptoms, indicating a possible measurement bias for Black Americans. The items were Item 1 (feeling nervous, anxious, or on edge), 5 (being so restless that it is hard to sit still), and 6 (becoming easily annoyed or irritable). This same study reported a lower prevalence of GAD among Black/African American participants but suggested that it was due to a bias in the instrument rather than a true difference in GAD between racial/ethnic groups. This could mean a lower cutoff of the GAD-7 is necessary to accurately identify Black persons with anxiety symptomatology.

Racial and Ethnic Differences in Diagnosis of Mental Health Conditions

Evidence indicates that the risk of misdiagnosis is elevated for Black and Latinx Americans. Data from 1995-2005 indicated that African American and Hispanic primary care patients were less likely to be diagnosed with depression or anxiety when compared with White patients.²²⁷ It is difficult to determine the reason for this difference in prevalence. Evidence from the 1999 National Health Interview Survey indicated that requiring a clinically significant impairment in order to receive a diagnosis of MDD led to an underestimate of depression for Black respondents.²²⁸ However we found very limited evidence exploring reasons for lower prevalence of depression in Black patients or those in other race and ethnic groups, compared with White patients.

Examples of differences in depression presentation were described above, which may lead to underdiagnosis with a presentation considered “atypical” in that it does not present primarily with sadness or depressed mood. A study with Black and White actors simulating different depression presentations examined how licensed social workers and marriage and family therapists approached diagnosing depression.²¹⁵ This study found that diagnostic accuracy was high for both Black and White actors with a “typical” depression presentation (92-97% accuracy), but identification was lower for the “atypical” presentation (55-63% accuracy), regardless of whether the actor was Black or White. However, one study examining differential item functioning for 20 items on the National Comorbidity Survey found that the removal of items with differential functioning between Black, Latinx, and White participants did not change the pattern of prevalence differences between the groups.²²⁹

Black adults have a higher rate of being diagnosed with schizophrenia,²³⁰ a phenomena that has been documented across approximately 30 years. One group of researchers have found evidence to support a pattern of under-recognition of mood-related symptoms and over-emphasis of psychotic-spectrum symptoms, suggesting racial bias in the diagnosis of schizophrenia spectrum disorders that might also contribute to underdiagnosis of mood disorders.^231,232 For example, in a comparison of clinical and structured interview assessments, mood symptoms identified in the structured interviews were more commonly omitted in the clinical assessments of Black patients, compared with the assessments of White patients.²³² In a separate study, Black patients were more likely to be diagnosed as having schizophrenia spectrum disorders by unblinded assessors who used structured interviews than by expert diagnosticians who used transcripts of the same interviews but who were blind to patients’ race and ethnicity (i.e., race and ethnic cues were eliminated from transcripts).²³¹

In a separate line of research, data from the 1990s indicated that clinicians’ perception of the patient’s honesty proved to be a significant predictor of racial disparities in schizophrenia diagnoses.²³³ While this harmful bias may be expected to have changed over the intervening decades, unfortunately clinician bias appears to be a continuing issue. In a 2021 publication of implicit bias, medical students and psychiatrists were more likely to pair faces of Black individuals with words related to psychotic disorders (as opposed to mood disorders), non-compliance (as opposed to compliance), and antipsychotic medications (as opposed to antidepressant medications). White race and higher level of training were the strongest predictors of associating faces of Black individuals with psychotic disorders, even after adjusting for the clinician’s age.²³⁴

Although diagnosis of children is beyond the scope of this review, another area of documented misdiagnosis is ADHD. In contrast to overdiagnosis of schizophrenia, ADHD in Black and Latinx children appear to be underdiagnosed.²³⁵ Compared with White children and controlling for ADHD symptoms and numerous sociodemographic variables, Black children had a 60% lower odds (OR, 0.60; 95% CI, 0.27 to 0.59) and Latinx children had a 63% lower odd (OR, 0.57; 95 % CI, 0.22 to 0.60) of having received an ADHD diagnosis than White children.

Racial and Ethnic Differences in Mental Health Referral and Service Use

According to 2008-2012 NSDUH data, the rates of mental health service use are higher for US residents who are White (16.6%), multiracial (17.1%), and Native American or Alaska Native (15.6%) than people who are Black (8.6%), Latinx (7.3%) and Asian American (4.9%). There is a similar pattern across race/ethnic groups for any use of outpatient mental health services and psychotherapeutic medications. This pattern also held for men and women and across age groups and poverty status.²³⁶

Similarly, among persons meeting criteria for anxiety, depression, alcohol or substance use disorder, or PTSD, the Collaborative Psychiatric Epidemiologic Studies (CPES) found that White persons had the highest rates of having at least one mental health visit (35.4%), followed by Latinx (30.6%), Black (28.8%), and Asian Americans (22.9%).²³⁷ Even controlling for the sociodemographic variables and insurance status, Asian and Black Americans had lower odds of having at least one visit compared to White Americans (OR, 0.56 [95% CI: 0.39, 0.82] and 0.74 [95% CI: 0.60, 0.92], respectively). Hispanic participants did not differ from White Americans. Asian and Black Americans had similar rates of visits to mental health practitioners in non-medical and spiritual settings that other race and ethnic groups, however.

In a group of not-for-profit healthcare systems, Asian, Black, Hispanic, Native Hawaiian or other Pacific Islander, Native American or Alaskan Native, and multiracial patients were less likely to be diagnosed with an anxiety disorder than White patients, and all of these groups except Native American or Alaskan Native patients were less likely to be diagnosed with depression. Among those with a mental health diagnosis, 77.8% of Non-Hispanic white patients received psychotropic medication, which was higher than other racial and ethnic groups (range 61.5% to 74.0%).²³⁸ On the other hand, several racial and ethnic groups other were more likely to received psychotherapy (e.g., 33.4% among White patients, 35.7% among Black patients, 39.5% among Native American or Alaskan Native patients).

Unfortunately, the disparity in mental health treatment among people from different race and ethnic groups widened between the years 2004 and 2012.²³⁹ Using data from a large nationally representative sample based on the Medical Expenditure Panel Surveys, the disparity widened between White respondents and Black, Latinx, and Asian American respondents in the proportion receiving any mental health care (from 8.2% to 10.8% difference between groups) and any psychotropic medication (from 7.6% to 10.0% difference). We were unable to find more recent data to determine whether this trend has continued.

Across all race and ethnic groups, cost or lack of insurance were listed as the main reason for not using mental health services, among those with a need for services.²³⁶ Cost is an important barrier for psychiatric care because the percentage of psychiatrists who accepted Medicare is historically lower than other physicians (e.g., 54.8% vs 86.1% in 2009-2010).²⁴⁰ The impact of cost hits families of color particularly hard, given the numerous federal policies that have disadvantaged Black and other people of color with respect to accumulation of wealth within and across generations. Cost and lack of insurance was followed by structural barriers for most race and ethnic groups (e.g., lacking transportation, not knowing where or how to access care). Few persons in any race and ethnic group reported low confidence that mental health services would help, but this reason was endorsed most by Asian (12.9%) and White (9.7%) respondents, followed by multiracial (7.2%), Latinx (6.5%), Black (5.3%), and Native American/Alaska Native (4.4%) respondents.²³⁶

Racial and Ethnic Differences in Treatment Effectiveness

Determining factors that promote success in specific populations is important. A study using the Collaborative Psychiatric Epidemiology Surveys found that Black, Latinx, and Asian American patients who were seen by a mental health specialist (vs a primary care provider) and were prescribed medication (vs therapy alone) were more likely to remain in treatment.²⁴¹ This study found that having a mental health specialist (vs a primary care provider) resulted in the greatest impact on treatment retention. One study hints that online interventions may also help narrow gaps in treatment benefits between White and Black patients with anxiety or depression. A recent study of 2,884 patients with elevated mood and/or anxiety symptoms found the intervention to have a larger benefit for Black participants than White participants.²⁴² This intervention used Internet-delivered interactive sessions and “homework” assignments to complete between weekly sessions, and was supported by bachelor-level care coaches.

An overview of reviews on the effectiveness of culturally-adapted mental health interventions found that effect sizes in 12 included reviews ranged from 0.23 to 0.75, with the majority reporting moderate to large effect sizes.²⁴³ In this review, adaptations were related to language, context, concepts, family, communication, content, cultural norms and practices, context and delivery, therapeutic alliance, and treatment goals. The reviewers noted some important limitation to this evidence, however, including a “lack of standardized frameworks for cultural adaptation of interventions that have been universally accepted, evaluated, and applied routinely in research and practice.”

A recent review covering a wide range of mental health conditions identified 57 RCTs reporting an effect on symptom reduction when adapted interventions were compared to non-adapted active treatments (K = 30, Hedge’s g = −0.43 [95% CI: −0.61, −0.25], p < .001).²⁴⁴ In this review, organization-specific adaptations had even larger effects than therapist- or content-specific adaptations in meta-regression (p=.02); organization-specific adaptations included the format used to provide treatment (e.g., face-to-face, digital), the location of treatment (home, community, non-healthcare), the contact time and length of treatment, and method of access (accelerated access, access via alternative to standard routes).

Mitigating Inequities in Mental Health Among Race and Ethnic Groups

While increased access to care for race and ethnic minority populations is important, straightforward expansion of services as they currently exist are unlikely to eliminate inequities in mental health. A recent review proposed the following strategies (and rationales for each) to help achieve mental health equity: payment reform and increasing population-based care; delivery system reform, increasing community-based health care services; addressing the social determinants of health; engaging local communities; enhancing the pipeline of clinicians from diverse backgrounds; and supporting a diverse, structurally competent workforce.²⁰⁰ Specific strategies for improving the healthcare delivery system include reducing barriers to access to care (e.g., increased availability of mental health services in geographic areas that are underserved, ease of referral from primary care settings), and increasing the use of (and training in the use of) integrated care settings. Engaging affected communities would enable the development of services that respond to patients’ needs and preferences and promote flexibility in evidence-based practice.²⁴⁵

The American Psychological Association issued the following recommendations to help eliminate inequities in mental health status and mental health care through the use of psychological and behavioral research and services that are culturally and linguistically competent:²⁴⁶

• Facilitate partnerships among physicians, mental and behavioral health providers, educators, community leaders, government agencies, and families to ensure development and implementation of culturally and linguistically competent and evidence-based prevention, early intervention, and treatment.
• Increase the availability of culturally and linguistically competent mental and behavioral health services accessible to racial and ethnic minorities.
• Increase research examining the complexities and intersections of multiple statuses/identities (e.g., socioeconomic status, disability, and immigrant status) and how these may contribute to psychological health.
• Foster positive relationships and programs within racial and ethnic minority communities to increase awareness of mental health issues and prevent environmental factors that may place individuals at risk.
• Increase funding for training mental and behavioral health professionals and to train these professionals to become culturally and linguistically competent.
• Develop and implement policy and programs based on psychological and behavioral research ensuring that racial and ethnic minorities are empowered through culturally and linguistically informed and evidence-based strategies.
• Advocate for local, State and national funding agencies to incorporate culturally and linguistically competent guidelines into proposals for programs for racial and ethnic minority children, youth, and families.
• Increase collaboration across federal funding organizations involved in racial and ethnic minority resiliency research.