PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
J Health Care Poor Underserved. Author manuscript; available in PMC 2011 May 1.
Published in final edited form as:
PMCID: PMC2874983
NIHMSID: NIHMS191152

Assessing Racial/Ethnic Differences in the Social Consequences of Early-Onset Psychiatric Disorder

Abstract

Individuals with early onset of psychiatric disorder have worse social outcomes than individuals with adult onset. It is unknown whether this association varies by racial/ethnic group. Identifying groups at risk for poor social outcomes is important for improving clinical and policy interventions. We compared unemployment, high school dropout, arrest, and welfare participation by race/ethnicity and time of onset using a nationally representative sample of Whites, Blacks, Asians, and Latinos with lifetime psychiatric disorder. Early onset was associated with worse social outcomes than adult onset. Significant Black-White and Latino-White differences in social outcomes were identified. The association between early onset and negative social outcomes was similar across Whites, Latinos, and Blacks. For Asians, the association between unemployment and early onset was opposite that of Whites. Increasing early detection and treatment of psychiatric illness should be prioritized. Further study will clarify the association between onset and social outcomes among sub-ethnic populations.

Keywords: Social determinants, mental health, racial/ethnic disparities, children’s mental health

Previous studies have found that individuals with early onset of psychiatric disorder (at or before eighteen years of age) have worse social outcomes as adults than people with adult onset of psychiatric disorders.1-3 More specifically, those with early-onset affective disorders experience worse employment and greater welfare dependence than those with late onset.1 Additionally, those with early onset of substance abuse achieve less education and employment opportunities than those with later onset or no disorder.2, 3 Finally, people with early onset of anxiety disorders have decreased likelihood of employment.2

Other authors have argued that, in a process of social selection, individuals with mental illness end up living in worse social conditions due to lack of self-care or underemployment.4, 5 Living in poor social conditions may contribute to a downward drift in society as the disadvantages of poor social conditions and psychiatric disorders accumulate over time.6 By adulthood, people with early onset of disorder may thus experience more of these stressors than those whose psychiatric disorders surface later in life. Given that half of individuals with a lifetime psychiatric disorder have initial onset of illness during childhood or adolescence,7 identifying subgroups that are particularly at risk for this downward trajectory is important for developing future clinical and policy interventions.

Prior studies have not assessed whether the association between early onset of psychiatric disorders and negative social outcomes varies by race/ethnicity. However, this variation might be expected given the racial/ethnic differences in both the prevalence of certain risk factors (e.g., Blacks have lower levels of education than Whites8-10) and the differing impact, or returns on these risk factors (e.g., the negative social consequences of low levels of education for Blacks are worse than those for Whites11, 12).

One of these important potential risk factors is racial/ethnic differences in education. Because of the lower levels of academic achievement among Black and Latino youth compared to White youth,8-10 it may be expected that Blacks and Latinos with early-onset disorder may suffer worse social outcomes than Whites with similar onset. For example, in every grade level, Blacks perform more poorly on standardized tests than Whites8,9 and Blacks and Latinos are less likely than Whites to graduate from high school.10 These differing levels of education are especially important in this context given that, among those with early onset-illness, truncated education is an important predictor of poor social outcomes.3 Poorer educational outcomes have been found among those with depression,13 trauma (particularly among minority youth),14 and attention deficit-hyperactivity disorder.15-17 On the basis of these studies, we hypothesize that Blacks and Latinos with early onset of psychiatric disorder will have lower levels of education than Whites with early onset of disorder, and thus worse outcomes in employment and other social domains.

A second potential risk factor is poorer access to mental health treatment given that Black and Latino youth are less likely to receive treatment for psychiatric disorders than Whites,18-23 and continue to be less likely to receive treatment across the lifespan,24,25 leading to a greater persistence of psychiatric disorder.26 Untreated psychopathology is, in turn, likely to be associated with negative social outcomes, particularly through its unfavorable effects on academic achievement.27,28 Because Blacks and Latinos are less likely to receive mental health care, we hypothesize that early onset of disorder will be especially detrimental to their social outcomes later in life.

As mentioned above, there may also be differential impact (or, returns) of education29 and mental health services use30,31 in their effects on the relationship between early onset disorders and negative social outcomes for Blacks and Latinos compared with Whites. That is, having greater number of years of education may not confer the same benefits for ethnic and racial minorities if the quality of their education is lower than Whites. This is a strong possibility given that racial/ethnic minorities are more likely to attend schools in segregated areas with lower quality instruction and higher levels of disorder and violence.32 There is also evidence of lower health returns on education among immigrants receiving education outside of the United States,33 a finding that is likely to disproportionately impact Latino-white differences. Likewise, the same intensity of mental health or substance abuse services may lead to poorer outcomes if the care received by racial/ethnic minorities is of lower quality, as has been shown among ethnic and racial minorities who live in disadvantaged areas.34 For these reasons, it may be important to assess how education and mental health treatment mediate the relationship between onset of mental disorder and later social outcomes.

In this study, we assess the association of early onset of psychiatric disorders, rather than adult onset, with later employment, high school graduation, arrest history, and public assistance program participation across major racial/ethnic groups in the United States. We hypothesize this association will be greater for Latinos and Blacks than it is for Whites. In addition to treating dropout from high school as a negative social outcome, we also test whether this measure of education is a significant mediator of the relationship between race/ethnicity, age of onset, and negative social outcomes. Lower levels of education and reduced access to mental health treatment are tested as plausible mediators of the association between race/ethnicity, age of onset, and negative social outcomes. Asians, as an aggregate group, tend to be more similar than other minorities to Whites in terms of income, education, and prevalence of mental health disorders35 (albeit that Asians attain more education than any other racial/ethnic group in the United States36). We thus hypothesize that the association between onset and negative social outcomes in adulthood will be similar for Whites and Asians.

Methods

Data

We used data from the Collaborative Psychiatric Epidemiological Surveys (CPES), a combination of the National Latino and Asian-American Study (NLAAS) dataset for Latinos and Asians, the National Comorbidity Survey Replication (NCS-R) dataset for Whites, and the National Survey of American Life (NSAL) for African Americans and Afro-Caribbeans. These three surveys used identical epidemiological instruments to assess mental disorders and service use, and provide a representative sample of English, Spanish, Tagalog, Vietnamese, and Chinese-speaking household residents ages 18 and older in the non-institutionalized population of the coterminous United States. The University of Michigan Survey Research Center collected data for all CPES datasets via in-person household interviews or telephone. National Latino and Asian-American Study data were collected during 2002-2003 and had a weighted response rate of 75.5% for Latinos and 65.6% for Asians.37,38 The National Comorbidity Survey Replication data were collected from 2001-2002 and had a weighted response rate of 70.9%.39 The National Survey of American Life data were collected between 2001 and 2003 and had a 72.3% response rate.40 Race and ethnicity (non-Latino White, Black, Latino, and Asian)* were ascertained using self-report responses to questions identical to those used in the 2000 Census. Individuals of any race claiming to be of Hispanic origin were identified as Latino in our study. Other respondents were classified as Black, Asian, or non-Latino White on the basis of responses to the question about race. Sampling weights were used so that results are representative of the national, non-institutionalized adult population.

We sub-sampled those individuals with any lifetime psychiatric disorder (including any lifetime substance abuse or dependence) for a total sample of 5,839 Whites, Blacks, Latinos, and Asians. We identified those with lifetime psychiatric disorder using self-report diagnostic measures from the diagnostic interview of the World Mental Health Survey Initiative version of the World Health Organization Composite International Diagnostic Interview (WMH-CIDI),41 a structured diagnostic instrument based on criteria of the DSM-IV.42

Along with race/ethnicity, the other independent variable of interest was age of onset, coded as early onset [age 0-18 years] or adult onset [age greater than 18 years]. Age of onset was determined for every individual diagnosed with a psychiatric disorder by asking the exact age at which the first episode of the psychiatric disorder occurred. It was considered to be the earliest date of onset of the earliest recorded psychiatric disorder, assessed using a series of questions. The first question was, “Can you remember the exact age the very first time you [HAD THE SYNDROME]?” Respondents answering “no” were asked to narrow down a time period by moving up in age categories, “Was it before you started school?” “Was it before you became a teenager?” For each category, age of onset was set at the upper limit of the category (e.g., age of onset was set to 12 for those answering yes to having age of onset before they became a teenager).

We used four dependent variables: (1) having less than a high school education (compared with high school education or greater); (2) ever arrested or convicted of a crime; (3) ever participated in a welfare or public assistance program; and (4) being unemployed (compared with having full- or part-time employment). For the fourth variable, we excluded all individuals who reported they were out of the work force and over the age of 65, in order to exclude individuals likely to be retired. We selected these four outcomes because they have serious and costly social consequences and because of evidence that they might be linked to the occurrence of psychiatric disorders.17,43-47

We adjusted for age category (18-24, 25-34, 35-44, 45-54, 55-64, 65 years old or older) in order to control for the possibility that younger cohorts had less time at risk of having an adult age of onset than older cohorts had. We additionally adjusted for gender and marital status because of their significant associations with socioeconomic status, arrest record, and welfare participation.48-51 These adjustments isolate the independent effects of early onset of psychiatric disorder and race/ethnicity on the four negative social outcomes. We test the mediating effects of education and mental health treatment using an indicator of high school dropout (compared with high school education or greater), and any lifetime mental health visit, which includes any mental health visit with a psychiatrist, psychologist, social worker or counselor, or mental health visits with a general practitioner, other medical doctor, nurse, occupational therapist, or other health professional for a mental health problem.

Statistical analyses

First, we present descriptive characteristics of the CPES subsample data, including unadjusted rates of last year employment, having less than high school education, arrest history, and welfare or public assistance program participation by age of onset and racial/ethnic category. We also describe age of onset and racial/ethnic differences in age, sex, and marital status, variables used as controls in the regression described below.

Next, we estimated logistic regression models for each of these dependent variables, adjusting for age, gender, and marital status. We examined the interaction of the racial/ethnic categories with age of onset in order to account for diverse racial/ethnic by age-of-onset group effects. Main effects of variables involved in interaction terms were centered by subtracting each response from the variable mean in order to make regression coefficients directly interpretable.52 For example, by centering, we were able to interpret the Black coefficient as the effect of Black race (compared with White) given an average age of onset. If we did not center, we would only be able to interpret the Black coefficient as the effect of Black race given adult onset of disorder (the referent group of the early-onset variable). Interaction coefficients were used to improve model fit and to estimate predicted probabilities of each race/ethnicity by age of onset subgroup (see below) but are not presented because, in non-linear models, they do not represent the marginal effect of the interaction term.53 The fit of these logistic models was verified using Pregibon’s Link Test54 and the Hosmer-Lemeshow goodness of fit test modified to account for intra-cluster correlated data and design effects.55

Third, to assess the mediating influence of education, we assessed the change in the coefficient of interest (in this case the interaction of race/ethnicity and age of onset) between a model with and without an indicator of having less than a high school education.56 We also conducted a Sobel test,56-58 an additional test of mediation that incorporates the association between the independent variable (race/ethnicity by age-of-onset interaction) and the mediator as well as the association between the mediator and the dependent variable (e.g., any lifetime arrest), adjusting for all other variables. The same mediation analyses were repeated for testing the significance of the mediating effect of any lifetime mental health treatment.

Fourth, we predicted rates of the four dependent variables for each race/ethnicity group by age-of-onset group, adjusting for all covariates. We applied the model coefficients from the model (described above) to the population as if they were all White with an adult onset of psychiatric disorder. We then repeated the prediction process for all other race/ethnicity by age of onset combinations: Black/adult onset, Latino/adult onset, Asian/adult onset, White/early onset, and so on. In total, we generated eight predictions for each dependent variable (four racial/ethnic groups by two age-of-onset categories), and then compared the average predictions among groups. Using these predictions, we illustrate the effect of age of onset by presenting bar graphs depicting minority-White differences in rates of negative social outcomes by age of onset categories.

This technique, named recycled predictions59, 60 or predictive margins,61 has been used in previous health services studies62-64 and allows us to make comparisons between ethnicity/race by age-of-onset groups, after standardization for all other variables. This method is a generalization of the adjusted treatment means to nonlinear models, allowing us to compare the rates of negative social outcomes of racial/ethnic groups with different age of onset of psychiatric disorders, after adjusting the distribution of all other observable attributes as if they were equal.

All analyses used sampling weights to provide estimates that are representative of the U.S. adult population with a lifetime history of psychiatric disorder. Standard errors for Tables Tables11 and and22 were estimated using Stata 10 software59 which accounts for the survey sampling design, and significance tests were performed using design-adjusted Wald tests. For recycled predictions, we derived standard errors from a bootstrap procedure65 and considered predicted differences significant if their 95% bootstrap intervals did not include zero.

Table 1
Unadjusted weighted characteristics of adult population with any lifetime psychiatric disorder. n=5,838
Table 2
Regression Coefficients from models fit to assess relationship between social position indicators and age of onset, race/ethnicity, age, sex, marital status (n=5,838)

Results

Table 1 describes our data, comparing unadjusted differences in negative social outcomes and predictors by age of onset and race/ethnicity among the population with any lifetime psychiatric disorder. Individuals with early onset of disorder were more likely to have a lifetime record of arrest than those with adult onset. They were also more likely than the adult-onset group to be unmarried. Blacks and Latinos with a lifetime psychiatric disorder were more likely than their White counterparts to be unemployed, to have dropped out of high school, and to have been arrested. Blacks with lifetime psychiatric disorder were also more likely than Whites to have ever participated in welfare or other public assistance programs. Asians with lifetime psychiatric disorder were more likely than Whites to be unemployed. In general, racial/ethnic minorities with lifetime psychiatric disorder were younger than Whites. Compared with Whites with lifetime psychiatric disorder, Latinos and Asians with lifetime psychiatric disorder were less likely to be widowed/ separated/ divorced and more likely never to have married, and Blacks were less likely to be married.

Multivariate results

After adjustment for race/ethnicity, age, sex, and marital status, we found individuals with early onset of psychiatric disorders were more likely to have less than a high school education, to have an arrest record, and to have been on welfare or another public assistance program (Table 2). After adjustment for age of onset, age, sex, and marital status, Latinos were more likely than Whites to be unemployed, to fail to complete high school, and to have been arrested. Asians were more likely to be unemployed, less likely to be arrested, and less likely to be on welfare, and Blacks were significantly more likely to experience all four of the measured negative social outcomes. As can be seen in Table 2, age, sex, and martial status were significant predictors of nearly all of the four negative social outcomes.

Mediation analysis results

Adding education to the models of unemployment, prior arrest, and welfare participation decreased the coefficients on the race/ethnicity by early-onset interaction variables, but not significantly. There is no conclusive evidence of a mediating effect of education on the relationship between race/ethnicity, early onset of disorder, and social outcomes (full results of separate mediation analysis available from authors upon request). This was verified by a Sobel test of mediation which found that though education was a significant mediator of the main effect of age of onset, it did not significantly mediate the relationship between the interaction of racial/ethnic group by age-of-onset and the negative social outcomes. Any lifetime mental health treatment was found to be an insignificant mediator of the relationship between the main effect of age of onset and negative social outcomes, as well as the relationship between the interaction of age-of-onset and racial/ethnic group and negative social outcomes.

Predictions of negative social outcomes by age of onset

For the total population with lifetime psychiatric disorder, individuals with early onset of psychiatric disorder were more likely than those with adult onset to be unemployed, to have less than a high school education, to have a lifetime arrest, and to have been enrolled in welfare or another public assistance program (Figure 1).

Figure 1
Predicted Rates of Social Indicators by Age of Onset

Table 3 presents our main results, which are also presented graphically in Figure 2. Whites and Blacks with early onset were more likely to be unemployed than those with adult onset, but the difference in age of onset compared across the two racial/ethnic groups (difference-in-difference) was not significant. The difference-in-difference for Asians versus Whites was significant. Adult-onset Asians were more likely to be unemployed than early-onset Asians, whereas adult-onset Whites were less likely to be unemployed than early-onset Whites, resulting in a racial/ethnic difference in onset differences of nearly 10%.

Figure 2
Assesing the Effect of Age of onset on Minority-White Differences in Negative Social Outcomes
Table 3
Predicted Rates of Social Position Indicators Race and Age of Onset.

Whites and Blacks with early onset were more likely to have less than a high school education than Whites and Blacks with adult onset. Whites and Asians with early onset were more likely to have ever been arrested than Whites and Asians with adult onset, and Blacks with early onset were more likely to have participated in welfare or other pubic assistance programs than Blacks with adult onset. Latinos with early onset had higher rates on the four negative social outcomes than Latinos with adult onset (e.g., Latinos with early onset were 5% more likely to have been arrested than Latinos with adult onset), but these differences were not significant. No significant difference-in-differences were found for high school graduation, arrest, or public assistance participation.

Discussion

Our assessment of racial/ethnic differences in the social consequences of early onset of psychiatric disorder has three main findings. First, we found that having an early onset of psychiatric disorder is more detrimental to social function than having adult onset for both the overall population and within White, Black, and Latino subpopulations. Compared with adult onset, we found early onset was associated with an increased probability of arrest, unemployment, participation in public assistance, and high school dropout. These findings complement previous work that supports a pathway from early onset of psychiatric disorder to negative social consequences.1-3 Our conclusions provide support for aggressive efforts promoting the timely identification and treatment of psychiatric disorders in childhood. Potential gains include not only alleviation of emotional suffering but also improvements across a range of important social domains.

Second, analysis of the interactions between Black and Latino race/ethnicity and age of onset suggests that the association between early onset and poor social outcomes is similar across White, Latino, and Black groups, and that this association was not amplified for Blacks and Latinos as we hypothesized. We also hypothesized that there would be racial/ethnic differences in the degree to which education and mental health treatment mediate the relationship between early age of onset and negative social outcomes. This was not supported by the data. The mediation of education was significant on the main effect between early age of onset and unemployment, welfare participation, and prior arrest, but did not differ by race/ethnicity. Mental health treatment was not found to be a significant mediator either for the main effect of age of onset or of the interaction between race/ethnicity and age of onset. This provides preliminary evidence that the higher rates of negative social outcomes among racial/ethnic minorities with early onset of psychiatric disorders compared with Whites operate through higher prevalence of low education in racial/ethnic minorities rather than through a differential effect of education on the relationship between early onset of psychiatric disorders and negative social outcomes.

There were significant first differences between Blacks and Latinos compared with Whites in all four of the negative social outcomes for both early- and adult-onset. For example, among those with early onset of psychiatric disorder, we found that 37% of Blacks and 41% of Latinos were unemployed, compared with 29% of Whites, and that high school dropout rates were 31% for Blacks and 40% for Latinos, compared with 15% for Whites. Differences in these rates of social disadvantage mirror those in the general U.S. population,66,67 except that the rates are significantly elevated (two to six times as high) for those with psychiatric disorder. Racial/ethnic differences in these negative social outcomes may be driven by the cumulative negative health effects of the social and economic adversity and political marginalization experienced by Blacks and Latinos in the U.S.68-70 These processes may have negative effects on physiological processes, which in turn have been linked to socioeconomic status71,72 and employment.73 The downward pressure on health and social status inflicted by these increases in allostatic load69, 70 may be especially detrimental to those with psychiatric illness.

Our third major finding was that the positive relationship between early onset and negative social outcomes was not seen in the Asian-White comparisons. Among Asians with a history of psychiatric disorder, there was no relationship between early onset and educational limitations or public assistance participation, and a negative relationship between early onset and unemployment. We suggest several possible reasons for this pattern. First, it has been found that Asians (both U.S.- and foreign-born) report psychiatric symptoms at a threshold of higher severity.74 Therefore, cases may have been misclassified as adult onset despite the emergence of significant psychiatric symptoms earlier in life, leading us to under-estimate the effect of early onset disorder on later social functioning. Second, the Asians in this sample immigrated to the United States later (mean age = 26.7 years) than other immigrant groups (mean age for Latinos = 20.8 years). Acculturative stress after migration, a risk factor for psychiatric disorder, may manifest in adulthood among Asian immigrants, which would also increase the association between adult-onset disorder and poor social outcomes.

In addition to the impact of misclassification or migration, Asians with early onset of illness have higher levels of education, which may be protective against unemployment, criminality, and welfare use.75 Asian families place great importance on education, and as a result they are the most highly educated cultural group in the United States.36,76 This finding was also reflected in our Asian sample’s low high school dropout rates. Fourth, these differences may be due to higher levels of mental health treatment among the early-onset group. Asians with early onset were more likely to be U.S.-born, and U.S.-born Asians are more likely to receive mental health care than foreign-born Asians.77 Furthermore, because illness is seen as an impediment to academic success, mental health treatment may be accepted more readily by parents. There are supporting data suggesting that Asian American adolescents remain in out-patient mental health treatment longer than Whites.78 Finally, there are important barriers to mental health treatment77 that may be more salient for Asian American adults than youth. These barriers relate to cognitive processes (e.g., failure to identify personal problems or emotional distress as mental illness worthy of treatment), affective issues (e.g., shame or stigma), or cultural value differences (e.g., collectivist values may conflict with the individual orientation of psychotherapy) related to mental illness and shared by Asian Americans.79

One reason for not finding significant interaction between racial/ethnic group and age of onset may be due to our limited sample size. Difference-in-difference analyses require a large sample to identify significant variations among the age-of-onset groups within each racial/ethnic group. A larger sample size is also needed to assess how the relationship between age of onset and negative social outcomes varies by race/ethnicity and type of psychiatric disorder. Anxiety and impulse disorders have a median age of onset of eleven years while mood and substance use disorders occur in later adolescence and adulthood.80 The distribution of these disorders was comparable across racial/ethnic groups in our sample (results not shown), suggesting that our main findings are not masking important differences in prevalence across racial/ethnic groups. However, important racial/ethnic differences in the effect of these separate disorders on negative social outcomes may be operating within our main findings. Sensitivity analyses run on a subsample of individuals with comorbid mental health and substance abuse disorders yielded significant, two- to three-fold increases in rates of lifetime arrest among those with comorbid substance and mental health disorder compared with those having only mental health disorder; however, these rates did not differ by age of onset. Furthermore, there was a greater differential impact of early onset of disorder on negative social outcomes for Blacks and Latinos with these comorbidities, although small sample sizes inflated standard errors and results were not significant. Future study on the differential impact of age of onset on negative social outcomes by race/ethnicity among individuals with comorbid substance and mental health disorders is warranted.

Another limitation of our data is that the CPES does not provide adequate sample size to assess differences in age of onset within Latino, African American, and Asian ethnic subgroups (e.g., Mexican, Puerto Rican, Cuban, Afro-Caribbean, Chinese, Vietnamese, Filipino) while adjusting for other covariates. Our prior research77, 81-83 found significant mental health and mental health service use differences between these sub-ethnic groups, suggesting that investigating the interaction of age of onset and sub-ethnicity would be worthwhile given a larger dataset. Unreliability of self-report data for age of onset is also a potential limitation for this study. This concern is mitigated by the methodological innovations introduced in the NCS-R, NLAAS, and NSAL, which encouraged active and extensive memory cuing to improve reliability of self-report estimates relative to earlier epidemiology studies.67

Our analysis of nationally representative data consistently identified an association between early age of onset of psychiatric disorder and negative social outcomes, implying that increasing childhood and adolescent psychiatric illness detection and treatment efforts should be prioritized. Blacks and Latinos had consistently higher rates of negative social outcomes than Whites within each age-of-onset category; however, the relationship between age of onset and these social outcomes appears to be the same as among Whites. These associations, especially for unemployment, were different for Asians, and we recommend future study to understand these patterns.

Footnotes

*For sake of brevity, these categories will be described as White, Black, Latino, and Asian from here forward.

REFERENCES

1. Fergusson DM, Boden JM, Horwood LJ. Recurrence of major depression in adolescence and early adulthood, and later mental health, educational and economic outcomes. Br J Psychiatry. 2007 Oct;191:335–342. [PubMed]
2. Turnbull JE, George LK, Landerman R, Swartz MS, Blazer DG. Social outcomes related to age of onset among psychiatric disorders. J Consult Clin Psychol. 1990 Dec;58(6):832–839. [PubMed]
3. Kessler RC, Foster CL, Saunders WB, Stang PE. Social consequences of psychiatric disorders, I: Educational attainment. Am J Psychiatry. 1995 Jul;152(7):1026–1032. [PubMed]
4. Loffler W, Hafner H. Ecological pattern of first admitted schizophrenics in two German cities over 25 years. Soc Sci Med. 1999 Jul;49(1):93–108. [PubMed]
5. Dembling BP, Rovnyak V, Mackey S, Blank M. Effect of geographic migration on SMI prevalence estimates. Ment Health Serv Res. 2002 Mar;4(1):7–12. [PubMed]
6. Kohn ML. Class, family, and schizophrenia: a reformulation. Social Forces. 1972:295–304.
7. Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR, Walters EE. Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry. 2005 Jun;62(6):593–602. [PubMed]
8. Campbell JR, Hombo CM, Mazzeo J. NAEP 1999 Trends in Academic Progress: Three Decades of Student Performance. 2000.
9. Neal DA. NBER Working paper. 2005. Why has Black-White skill convergence stopped?
10. Farkas G. Racial Disparities and Discrimination in Education: What do we know, how do we know it, and what do we need to know? The Teachers College Record. 2003;105(6):1119–1146.
11. Holzer HJ, Offner P, Sorensen E. What explains the continuing decline in labor force activity among young Black men? Labor History. 2005;46(1):37–55.
12. Holzer HJ, Offner P, Sorensen E. Declining employment among young Black less-educated men: The role of incarceration and child support. Journal of Policy Analysis and Management. 2005;24(2):329–350.
13. Fergusson DM, Woodward LJ. Mental health, educational, and social role outcomes of adolescents with depression. Am Med Assoc. 2002;59:225–231. [PubMed]
14. Lansford JE, Dodge KA, Pettit GS, Bates JE, Crozier J, Kaplow J. A 12-Year Prospective Study of the Long-term Effects of Early Child Physical Maltreatment on Psychological, Behavioral, and Academic Problems in Adolescence. Arch Pediatr Adolesc Med. 2002 August 1;156(8):824–830. 2002. [PMC free article] [PubMed]
15. Barkley RA, Anastopoulos AD, Guevremont DC, Fletcher KE. Adolescents with ADHD: Patterns of behavioral adjustment, academic functioning, and treatment utilization. Journal of the American Academy of Child and Adolescent Psychiatry. 1991;30(5):752–761. [PubMed]
16. Semrud-Clikeman M, Biederman J, Sprich-Buckminster S, Lehman BK, Faraone SV, Norman D. Comorbidity between ADDH and Learning Disability: A Review and Report in a Clinically Referred Sample. Journal of the American Academy of Child & Adolescent Psychiatry. 1992;31(3):439. [PubMed]
17. Hinshaw SP. Externalizing behavior problems and academic underachievement in childhood and adolescence: Causal relationships and underlying mechanisms. Psychological Bulletin. 1992;111(1):127–155. [PubMed]
18. Elster A, Jarosik J, VanGeest J, Fleming M. Racial and ethnic disparities in health care for adolescents: a systematic review of the literature. Archives of Pediatrics & Adolescent Medicine. 2003 Sep;157(9):867–874. [PubMed]
19. Garland AF, Lau AS, Yeh M, McCabe KM, Hough RL, Landsverk JA. Racial and ethnic differences in utilization of mental health services among high-risk youths. The American journal of psychiatry. 2005 Jul;162(7):1336–1343. [PubMed]
20. Kodjo CM, Auinger P. Predictors for emotionally distressed adolescents to receive mental health care. The Journal of Adolescent Health. 2004 Nov;35(5):368–373. [PubMed]
21. Kataoka SH, Zhang L, Wells KB. Unmet need for mental health care among U.S. children: variation by ethnicity and insurance status. The American journal of psychiatry. 2002 Sep;159(9):1548–1555. [PubMed]
22. Yeh M, McCabe K, Hurlburt M, et al. Referral sources, diagnoses, and service types of youth in public outpatient mental health care: a focus on ethnic minorities. Journal of Behavioral Health Services & Research. 2002;29(1):45–60. [PubMed]
23. Hough RL, Hazen AL, Soriano FI, Wood P, McCabe K, Yeh M. Mental health services for Latino adolescents with psychiatric disorders. Psychiatric services (Washington, D.C. 2002 Dec;53(12):1556–1562. [PubMed]
24. Jackson JS, Neighbors HW, Torres M, Martin LA, Williams DR, Baser R. Use of mental health services and subjective satisfaction with treatment among Black Caribbean immigrants: results from the National Survey of American Life. American journal of public health. 2007 Jan;97(1):60–67. [PubMed]
25. Neighbors HW, Caldwell C, Williams DR, et al. Race, ethnicity, and the use of services for mental disorders: results from the National Survey of American Life. Archives of general psychiatry. 2007 Apr;64(4):485–494. [PubMed]
26. Breslau J, Kendler K, Su M, Gaxiola-Aguilar S, Kessler R. Lifetime risk and persistence of psychiatric disorders across ethnic groups in the United States. Psychological Medicine. 2005;35(3):317–327. [PMC free article] [PubMed]
27. Scheffler RM, Brown TT, Fulton BD, Hinshaw SP, Levine P, Stone S. Positive Association Between Attention-Deficit/Hyperactivity Disorder Medication Use and Academic Achievement During Elementary School. Pediatrics. 2009;123(5):1273. [PubMed]
28. Fuligni AJ. The academic achievement of adolescents from immigrant families: The roles of family background, attitudes, and behavior. Child Development. 1997:351–363. [PubMed]
29. Barringer H, Takeuchi D, Xenos P. Education, Occupational Prestige, and Income of Asian Americans. Sociology of Education. 1990;63(1):27–43.
30. Chow J, Jaffee K, Snowden L. Racial/Ethnic Disparities in the Use of Mental Health Services in Poverty Areas. American Journal of Public Health. 2003;93:792–797. [PubMed]
31. Snowden L. Bias in mental health assessment and intervention: theory and evidence. American Journal of Public Health. 2003;93(2):239–243. [PubMed]
32. Massey DS. Social Background and Academic Performance Differentials: White and Minority Students at Selective Colleges. American Law and Economics Review. 2006;8(2):390–409.
33. Walton E, Takeuchi D, Herting JR, Alegría M. Contextualizing the Education and Health Status Association: Evidence from a National Study of Asian Americans. Biodemography and Social Biology. 2009;55:30–51. [PMC free article] [PubMed]
34. Galea S, Vlahov D. Urban Health: Evidence, Challenges, and Directions. Annual Review of Public Health. 2005;26(1):341–365. [PubMed]
35. United States Surgeon General Mental Health: Culture, Race, and Ethnicity: A Supplement to Mental Health: A Report of the Surgeon General. 2001. [PubMed]
36. U.S. Census Bureau 1990 profiles of Asians and Pacific Islanders: Selected characteristics. (CPH-L-151) U.S. Government Printing Office; Washington, D.C.: 1990.
37. Alegría M, Takeuchi D, Canino G, et al. Considering Context, Place and Culture: the National Latino and Asian American Study. International Journal of Methods in Psychiatric Research. 2004;13(4):208–220. [PMC free article] [PubMed]
38. Heeringa SG, Wagner J, Torres M, Duan N, Adams T, Berglund P. Sample designs and sampling methods for the Collaborative Psychiatric Epidemiology Studies (CPES) Int J Methods Psychiatr Res. 2004;13(4):221–240. [PubMed]
39. Kessler RC, Merikangas KR. The National Comorbidity Survey Replication (NCS-R): background and aims. Int J Methods Psychiatr Res. 2004;13(2):60–68. [PubMed]
40. Jackson JS, Torres M, Caldwell CH, et al. The National Survey of American Life: a study of racial, ethnic and cultural influences on mental disorders and mental health. International Journal of Methods in Psychiatric Research. 2004;13(4):196–207. [PubMed]
41. Kessler R, Ustun T. The World Mental Health (WMH) survey initiative version of the World Health Organization (WHO) Composite International Diagnostic Interview (CIDI) The International Journal of Methods in Psychiatric Research. 2004;13(2):93. [PubMed]
42. American Psychiatric Association Diagnostic and Statistical Manual of Mental Disorders, 4th edition-text revision (DSM-IV-TR) American Psychiatric Association; Washington, D.C.: 2000.
43. Abram KM, Teplin LA. Co-occurring disorders among mentally ill jail detainees: Implications for public policy. American Psychologist. 1991;46(10):1036–1045. [PubMed]
44. French M, Roebuck C, Alexandre P. Illicit Drug Use, Employment, and Labor Force Participation. Southern Economic Journal. 2001;68(2):349–368.
45. Frank RG, Glied SA. Better But Not Well: Mental Health Policy in the United States Since 1950. The Johns Hopkins University Press; Baltimore: 2006.
46. Turner L, Danziger S, Seefeldt K. Failing the transition from welfare to work: Women disconnected from work, welfare, and other sources of economic support. Social Science Quarterly. 2006;87(2):227–249.
47. Bray JW, Zarkin GA, Dennis ML, French MT. Symptoms of dependence, multiple substance use, and labor market outcomes. The American journal of drug and alcohol abuse. 2000;26(1):77–95. [PubMed]
48. Barrett AE, Turner RJ. Family structure and mental health: The mediating effects of socioeconomic status, family process, and social stress. Journal of Health and Social Behavior. 2005:156–169. [PubMed]
49. Fisher WH, Roy-Bujnowski KM, Grudzinskas AJ, Jr, Clayfield JC, Banks SM, Wolff N. Patterns and prevalence of arrest in a statewide cohort of mental health care consumers. Psychiatric Services. 2006;57(11):1623. [PMC free article] [PubMed]
50. White MC, Chafetz L, Collins-Bride G, Nickens J. History of arrest, incarceration and victimization in community-based severely mentally ill. Journal of Community Health. 2006;31(2):123–135. [PubMed]
51. Artazcoz L, Benach J, Borrell C, Cortes I. Unemployment and mental health: understanding the interactions among gender, family roles, and social class. Am Public Health Assoc. 2004;94:82–88. [PubMed]
52. Kraemer HC, Blasey CM. Centring in regression analyses: a strategy to prevent errors in statistical inference. Int J Methods Psychiatr Res. 2004;13(3):141–151. [PubMed]
53. Ai C, Norton E. Interaction terms in logit and probit models. Economics Letters. 2003;80:123–129.
54. Pregibon D. Goodness of link tests for generalized linear models. Applied Statistics. 1980;29:15–24.
55. Hosmer DW, Lemeshow S. Applied logistic regression analysis. John Wiley & Sons; New York: 1989.
56. Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology. 1986;51(6):1173–1182. [PubMed]
57. Mackinnon DP, Lockwood CM, Hoffman JM, West SG, Sheets V. A comparison of methods to test mediation and other intervening variable effects. Psychological Methods. 2002;7(1):83–104. [PMC free article] [PubMed]
58. Sobel ME. Asymptotic confidence intervals for indirect effects in structural equation models. Sociological Methodology. 1982;13:290–312.
59. StataCorp Reference Manual (Release 9.0) STATA Press; College Station, TX: 2005.
60. Kronick R, Gilmer T. Insuring Low-Income Adults: Does Public Coverage Crowd Out Private? Health Affairs. 2002;21(1):225–239. [PubMed]
61. Graubard B, Korn E. Predictive Margins with Survey Data. Biometrics. 1999;55(2):652. [PubMed]
62. Davern M, Rodin H, Blewett LA, Call KT. Are the Current Population Survey Uninsurance Estimates Too High? An Examination of the Imputation Process. Health Services Research. 2007;42(5):2038. [PMC free article] [PubMed]
63. Blewett LA, Davidson G, Bramlett MD, Rodin H, Messonnier ML. The Impact of Gaps in Health Insurance Coverage on Immunization Status for Young Children. Health Services Research. 2008;43(5p1):1619–1636. [PMC free article] [PubMed]
64. Wells KB, Tang L, Miranda J, Benjamin B, Duan N, Sherbourne CD. The Effects of Quality Improvement for Depression in Primary Care at Nine Years: Results from a Randomized, Controlled Group-Level Trial. Health Services Research. 2008;43(6):1952. [PMC free article] [PubMed]
65. Efron B. Bootstrap Methods: Another Look at the Jackknife. The Annals of Statistics. 1979;7(1):1–26.
66. National Center for Education Statistics The Condition of Education 2008: Table 23-1. 2008. http://nces.ed.gov/FastFacts/display.asp?id=16.
67. KnÄuper B, Cannell CF, Schwarz N, Bruce ML, Kessler RC. Improving accuracy of major depression age-of-onset reports in the US National Comorbidity Survey. International Journal of Methods in Psychiatric Research. 1999;8(1):39–48.
68. James SA, Keenan NL, Strogatz DS, Browning SR, Garrett JM. Socioeconomic Status, John Henryism, and Blood Pressure in Black Adults The Pitt County Study. American journal of epidemiology. 1992;135(1):59–67. [PubMed]
69. McEwen BS. Protective and damaging effects of stress mediators. New England Journal of Medicine. 1998;338(3):171–179. [PubMed]
70. Geronimus AT, Hicken M, Keene D, Bound J. “Weathering” and age patterns of allostatic load scores among Blacks and Whites in the United States. Am Public Health Assoc. 2006;96:826–833. [PubMed]
71. Marmot MG, Fuhrer R, Ettner SL, Marks NF, Bumpass LL, Ryff CD. Contribution of psychosocial factors to socioeconomic differences in health. The Milbank Quarterly. 1998:403–448. [PubMed]
72. Weinstein M, Goldman N, Hedley A, Yu-Hsuan L, Seeman T. Social linkages to biological markers of health among the elderly. Journal of Biosocial Science. 2003;35(03):433–453. [PubMed]
73. Steptoe A, Feldman PJ, Kunz S, Owen N, Willemsen G, Marmot M. Stress responsivity and socioeconomic status. A mechanism for increased cardiovascular disease risk? Eur Soc Cardiology. 2002;23:1757–1763. [PubMed]
74. Lin JY, Shrout P, Yang F, Alegria M. Measurement bias in the screener questions for the depression section of the Composite International Diagnostic Interview. Unpublished. 2009
75. Chatterji P, Alegria M, Lu M, Takeuchi D. National Bureau of Economic Research Working Paper 11893. 2005. Psychiatric Disorders and Labor Evidence from the National Latino and Asian American Study.
76. Jeynes WH. A Meta-Analysis: The Effects of Parental Involvement on Minority Children’s Academic Achievement. Education and Urban Society. 2003 February 1;35(2):202–218. 2003.
77. Abe-Kim J, Takeuchi D, Hong S, et al. Use of Mental Health-Related Services among Immigrant and US-born Asian Americans: results from the National Latino and Asian American Study. American Journal of Public Health. 2007;97(1):91–98. [PubMed]
78. Bui KVT, Takeuchi DT. Ethnic minority adolescents and the use of community mental health care services. American Journal of Community Psychology. 1992;20(4):403–417. [PubMed]
79. Leong FTL, Lau ASL. Barriers to providing effective mental health services to Asian Americans. Mental health services research. 2001;3(4):201–214. [PubMed]
80. Kessler R, Berglund P, Demler O, Jin R, Walters E. Lifetime Prevalence and Age-of-Onset Distributions of DSM-IV Disorders in the National Comorbidity Survey Replication. Archives of General Psychiatry. 2005;62(593-602) [PubMed]
81. Alegria M, Mulvaney-Day N, Torres M, Polo A, Cao Z, Canino G. Prevalence of psychiatric disorders across Latino subgroups in the United States. Am J Public Health. 2007 Jan;97(1):68–75. [PubMed]
82. Alegria M, Mulvaney-Day N, Woo M, Torres M, Gao S, Oddo V. Correlates of past-year mental health service use among Latinos: results from the National Latino and Asian American Study. Am J Public Health. 2007 Jan;97(1):76–83. [PubMed]
83. Jackson JS, Neighbors HW, Torres M, Martin LA, Williams DR, Baser R. Use of mental health services and subjective satisfaction with treatment among Black Caribbean immigrants: results from the National Survey of American Life. American Journal of Public Health. 2007 Jan;97(1):60–67. [PubMed]