Comorbid substance use disorders with other Axis I and II mental disorders among treatment-seeking Asian Americans, Native Hawaiians/Pacific Islanders, and mixed-race people

Abstract

Little is known about behavioral healthcare needs of Asian Americans (AAs), Native Hawaiians/Pacific Islanders (NHs/PIs), and mixed-race people (MRs)—the fastest growing segments of the U.S. population. We examined substance use disorder (SUD) prevalences and comorbidities among AAs, NHs/PIs, and MRs (N = 4572) in a behavioral health electronic health record database. DSM-IV diagnoses among patients aged 1–90 years who accessed behavioral healthcare from 11 sites were systematically captured: SUD, anxiety, mood, personality, adjustment, childhood-onset, cognitive/dementia, dissociative, eating, factitious, impulse-control, psychotic/schizophrenic, sleep, and somatoform diagnoses. Of all patients, 15.0% had a SUD. Mood (60%), anxiety (31.2%), adjustment (30.9%), and disruptive (attention deficit-hyperactivity, conduct, oppositional defiant, disruptive behavior diagnosis, 22.7%) diagnoses were more common than others (psychotic 14.2%, personality 13.3%, other childhood-onset 11.4%, impulse-control 6.6%, cognitive 2.8%, eating 2.2%, somatoform 2.1%). Less than 1% of children aged <12 years had SUD. Cannabis diagnosis was the primary SUD affecting adolescents aged 12–17. MRs aged 35–49 years had the highest prevalence of cocaine diagnosis. Controlling for age at first visit, sex, treatment setting, length of treatment, and number of comorbid diagnoses, NHs/PIs and MRs were about two times more likely than AAs to have ≥2 SUDs. Regardless of race/ethnicity, personality diagnosis was comorbid with SUD. NHs/PIs with a mood diagnosis had elevated odds of having SUD. Findings present the most comprehensive patterns of mental diagnoses available for treatment-seeking AAs, NHs/PIs, and MRs in the real-world medical setting. In-depth research is needed to elucidate intraracial and interracial differences in treatment needs.

Introduction

Substance use disorders (SUDs; alcohol- or drug-related disorders) take a heavy toll on the affected individuals and their families. Substance abuse hampers educational attainment and productivity, leads to criminal activities, affects almost all major organs of users, and results in premature mortality (Brick, 2008, Clark et al., 2008, Wickizer, 2013). An estimated 11% of U.S. youth aged 13–18 years have SUD in their lifetime, which is close to rates for oppositional defiant disorder (ODD; 13%), attention deficit-hyperactivity disorder (ADHD; 9%), and conduct disorder (CD; 7%) and lower than any anxiety (32%) or mood (14%) disorder (Merikangas et al., 2010). Among U.S. individuals aged ≥12 years, approximately 8% have had a SUD in the past year (SAMHSA, 2012a). Persons with SUD often have comorbid mental conditions, which may include internalizing (depressive), externalizing (AHDH), or personality disorders (Najt et al., 2011, O'Neil et al., 2011, Wu et al., 2011a). Occurrences of comorbidities are related to multiple mechanisms (e.g., self-medication of substance use for relieving mental stress, substance-related neurotoxic effects on mental health, shared genetic or environmental factors), and severity generally elevates with an increased number of diagnoses (Glantz et al., 2009, Swendsen et al., 2010). Treatment-use data from a large electronic health record (EHR) database demonstrate that patients with comorbid SUD and mental disorders are among the most severe subset of patients: they received more clinical diagnoses, had a longer length of treatment, and used more costly inpatient and emergency care than patients without comorbid SUD (Wu et al., 2011a, Wu et al., 2013a).

Due to costs, stigma, and fragmentation of the healthcare system, individuals with SUD often received inadequate care for their condition (Pating et al., 2012, Wu et al., 2011b). Passage of the Affordable Care Act is expected to improve healthcare for all medical conditions, especially SUD, for members of nonwhite communities (Pating et al., 2012). To inform strategies for improving healthcare, one step is to understand the extent of SUD and comorbidities affecting people in the real-world settings. During the past decade, there were major shifts in the racial/ethnic composition of nonwhite populations, particularly Asian Americans [AAs] (14.7 million; 4.8% of the U.S. population), Native Hawaiians/Pacific Islanders [NHs/PIs] (0.5 million; 0.2%), and mixed-race (MR) individuals or people of more than one race (9.0 million; 2.9%). These populations grew three times faster than the total U.S. population (U.S. Census Bureau, 2011). However, little is known about the extent of SUD and comorbidities in these growing populations.

The rising numbers of AAs, NHs/PIs, and MRs require empirical data to inform behavioral healthcare. To improve health statistics for these overlooked groups, the U.S. Census Bureau began in 2000 to report AAs, NHs/PIs, and MRs (Srinivasan and Guillermo, 2000). However, because of relatively small samples of these groups participated in a typical study, AAs and NHs/PIs are either pooled as “other” or “Asian” or excluded from analysis; MRs are infrequently reported. The Monitoring the Future report has not provided substance use estimates for these groups (Johnston et al., 2012). The Treatment Episode Data Set (TEDS) reports findings of substance abuse treatment admissions for pooled AAs/NHs/PIs and does not have estimates for MRs (SAMHSA, 2012a). In TEDS, alcohol, marijuana, and stimulants were the primary substances of abuse for pooled AAs/NHs/PIs (SAMHSA, 2012a), and the majority (>50%) of AAs/NHs/PIs were either not employed or unemployed (SAMHSA, 2010). National data show that the majority of NHs/PIs and MRs live in lower-income families than whites and AAs, suggesting a greater need for support to address SUD care (Wu et al., 2013b, Wu et al., 2013c).

Available substance use estimates for these groups rely mainly on adolescent data and lack information on SUD and comorbidity (Makimoto, 1998, Wallace et al., 2002). Prior data suggested prevalent rates of substance use among NHs/PIs or MRs (Edwards et al., 2010, Kim and McCarthy, 2006). Among 10th-graders, 52% of NHs used marijuana in their lifetime compared with 46% of whites (Wong et al., 2004). Among youths aged 16–23 years, MRs reported high prevalences of lifetime use of methamphetamine (MR: 11.4%, White: 6.1%, Black: 0.5%, Hispanic: 3.4%) and ecstasy (MR: 21.9%, White: 16.8%, Black: 4.2%, Hispanic: 8.9%) (Wu et al., 2006). Recent findings from people aged ≥12 years showed higher prevalences of alcohol, tobacco, and drug use among NHs/PIs and MRs than AAs (Wu et al., 2013b, Wu et al., 2013c). MRs had a higher 12-month prevalence of any drug use (21.1%) than Whites (14.9%), NHs/PIs (16.4%), and AAs (7.5%) (Wu et al., 2013b). However, there is a dearth of SUD data for these populations. One study determined SUD prevalences among adolescents (aged 12–17) and found higher 12-month prevalences of drug use disorders among MRs (6.4%) than Whites (5.0%), Blacks (3.7%), Hispanics (4.4%), and AAs/NHs/PIs (2.0%) (Wu et al., 2011c). Among adolescents who used alcohol or drugs in the past year, a high proportion (25.2%) of MRs had SUD (Whites 22.9%, Blacks 15.5%, Native-Americans 31.5%, AAs/NHs/PIs 14.9%, Hispanics 21.0%) (Wu et al., 2011c). These findings support the need to disaggregate SUD estimates for AAs and NHs/PIs and to include MRs in research.

Even less is known about comorbidities. The 2002–2003 National Latino and Asian American Study (NLAAS) showed that 3.6% (n = 71) of AAs had a lifetime alcohol disorder (Chae et al., 2008), and that 17.3% had any lifetime disorder, including substance, depressive, and anxiety disorders (Takeuchi et al., 2007). The NLAAS findings have not provided data for individual drug use disorders and comorbid SUD for AAs. The 2001–2002 National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) is presently the largest U.S. survey of comorbidities (N = 43,093) and includes 1332 non-Hispanic AAs/NHs/PIs (Huang et al., 2006). In NESARC, the 12-month prevalence of drug disorders (1.4%) and mood disorders (7.4%) in the pooled AA/NH/PI sample was similar to rates in Whites, Blacks, and Hispanics. In the pooled AA/NH/PI sample, lifetime alcohol disorder was associated with lifetime mood, anxiety, and personality disorders, and lifetime drug disorder was associated with lifetime mood disorder (Huang et al., 2006). There is no known finding on comorbidity among MRs from NESARC.

The limitation of survey-based studies indicates the need for other sources to examine SUD and comorbidities. To address the issue of sample size, we utilized the largest behavioral health EHR database in the United States to determine patterns of SUDs and comorbidities among AAs, NHs/PIs, and MRs. The analysis sample included patients aged 1−90 years who accessed any behavioral healthcare at 1 of 11 sites in the community. The Institute of Medicine (IOM) highlights the need to use patients' personal medical records (EHRs) for research to learn about treatment needs and healthcare delivery in real-world settings (IOM, 2010). Using the EHR database allows us to examine SUDs and comorbidities in geographically diverse groups of AAs, NHs/PIs, and MRs in such settings. We (1) examined prevalences of DSM-IV mental diagnoses comprehensively captured in the EHR database, (2) determined SUD prevalences and correlates of having one or more SUDs, and (3) estimated associations of comorbid diagnoses with SUD.