Abstract
Depression prevalence was examined by race/ethnicity in a nationally representative sample. The Diagnostic Interview Schedule was administered to 8449 (response rate=96.1%) participants (aged 15–40 years). Prevalence of major depressive disorder was significantly higher in Whites than in African Americans and Mexican Americans; the opposite pattern was found for dysthymic disorder. Across racial/ethnic groups, poverty was a significant risk factor for major depressive disorder, but significant interactions occurred between race/ethnicity, gender, and education in relation to prevalence of dysthymic disorder.
Published data regarding prevalence of depression by race/ethnicity are contradictory. Some studies report higher rates of major depressive disorder in African American compared with White individuals,1–3 whereas others report lower or equivalent rates in African American individuals.4–7 Few studies have examined dysthymic disorder by race/ethnicity.7 The purpose of this study was to examine the prevalence of major depressive disorder and dysthymic disorder in the general United States population by race/ethnicity.
METHODS
Survey Design
The National Health and Nutrition Examination Survey III was conducted from 1988 to 1994 by the National Center for Health Statistics.
Sampling followed a stratified, multistage, probability cluster design from which a representative US sample was obtained. Mexican American and African American individuals were oversampled for more reliable estimates. Respondents (N =8449; 96.1% response rate) aged 15 to 40 years were administered the Diagnostic Interview Schedule,8 which used criteria from the Diagnostic and Statistical Manual of Mental Disorders, Revised Third Edition (DSM-III-R ).9 Approximately 3.9% of the respondents had nonvalid or missing Diagnostic Interview Schedule data.
The 2 outcomes were (1) dysthymic disorder: at least 2 years of dysphoric mood (“[have you]…felt depressed or sad almost all the time, even if you felt OK sometimes?”) plus 2 other symptoms of depression, and (2) major depressive disorder: at least 2 weeks of depressed mood (“[have you]…felt sad, blue, depressed, or…lost all interest and pleasure in things that you usually cared about or enjoyed?”) plus 4 other symptoms.
Statistical Analyses
We used χ2 and logistic regression analyses (SAS, Version 8 [SAS Institute Inc, Cary, NC], and Stata, Version 7 [Stata Corp, College Station, Tex]) to assess whether racial/ethnic group was associated with prevalence of depression (by type), and we controlled for age, gender, income, education, and marital status. Sampling weights were used to correct for differential probability of selection and differential response rates by age, gender, and race/ethnicity.
RESULTS
Prevalence of major depressive disorder differed significantly by racial/ethnic group, with the highest prevalence in White participants (Table 1 ▶). Mexican American and White individuals had significantly earlier onset of major depressive disorder compared with African American individuals (P = .001). Overall, persons living in poverty had nearly 1.5 times the prevalence of major depressive disorder; however, poverty was significantly associated with prevalence of major depressive disorder only for White respondents (P = .023). Lack of education (< 8 years of school) was significantly associated with prevalence of major depressive disorder only for Mexican American individuals (P = .000).
TABLE 1—
Lifetime Prevalence and Logistic Odd Ratios (ORs) for Depression, by Type and Risk Category (N = 8449)
| Major Depressive Disorder a | Dysthymic Disorder b | |||||
| Risk Category | Prevalence | Unadjusted OR | Adjusted OR | Prevalence | Unadjusted OR | Adjusted OR |
| Race/Ethnicity | ||||||
| White | 10.40 | 1.00 | 1.00 | 5.70 | 1.00 | 1.00 |
| African American | 7.50** | 0.61** | 0.51** | 7.50* | 1.27* | 0.92 |
| Mexican American | 8.00* | 0.68** | 0.38** | 7.40* | 1.27* | 0.45* |
| Gender | ||||||
| Male | 6.30 | 1.00 | 1.00 | 7.70** | 1.00 | 1.00 |
| Female | 12.60** | 2.26** | 2.16** | 4.50 | 1.75** | 1.70** |
| Age, y | ||||||
| 15–19 | 7.50 | 1.00 | 1.00 | 4.70 | 1.00 | 1.00 |
| 20–24 | 10.20 | 1.23 | 1.25 | 4.40 | 1.31 | 1.72** |
| 25–29 | 8.30 | 1.07 | 1.10 | 5.70 | 1.67 | 1.60** |
| 30–34 | 9.60 | 1.23** | 1.32 | 7.70** | 1.81** | 2.74** |
| 35–40 | 11.60* | 1.43** | 1.46 | 7.80* | 1.73** | 2.60** |
| Income | ||||||
| Above poverty level | 9.10 | 1.00 | 1.00 | 4.90 | 1.00 | 1.00 |
| Below poverty level | 12.80** | 1.13 | 1.23* | 12.30** | 1.95** | 1.56** |
| Marital status | ||||||
| Never married | 8.70 | 1.00 | 1.00 | 5.60 | 1.00 | 1.00 |
| Married living with spouse | 8.80 | 0.99 | 0.81 | 5.00 | 0.98 | 0.80 |
| Separated, divorced, or widowed | 16.90** | 2.20** | 1.67** | 14.50** | 2.41** | 1.57** |
| Education, y | ||||||
| High school graduate | 9.70 | 1.00 | 1.00 | 6.30 | 1.00 | 1.00 |
| 0–8 | 8.90 | 0.73* | 0.85 | 12.10 | 1.43** | 1.65** |
| 9–11 | 9.70 | 1.01 | 1.10 | 8.20 | 1.37** | 1.69** |
| Some college | 9.50 | 1.17 | 1.17 | 3.80 | 0.76* | 0.77* |
| Total (overall prevalence) | 9.50 | … | … | 6.10 | … | … |
a The best model for major depressive disorder includes race/ethnicity (African American vs White, Mexican American vs White), gender (female vs male), income (below poverty level vs above poverty level), and marital status (married vs single; separated, divorced, or widowed vs single). Note that age and education were not significantly related to major depressive disorder in the combined model.
b The best model for dysthymic disorder includes race/ethnicity (African American vs White, Mexican American vs White), gender (female vs male), income (below poverty level vs above poverty level), marital status (married vs single; separated, divorced, or widowed vs single), age, and education.
*P < .05; **P < .01.
In contrast to the comparative rates for major depressive disorder, the prevalence of dysthymic disorder was significantly greater among African American and Mexican American individuals compared with Whites (Table 1 ▶). After we controlled for poverty, lack of education remained a significant risk factor for dysthymic disorder. In addition, significant interactions occurred between race/ethnicity, gender, and education in relation to prevalence of dysthymic disorder. Specifically, for White respondents (of both genders), a precipitous decline in prevalence of dysthymic disorder was seen with any education beyond middle school (> 8 years of education); however, for Mexican American and African American subjects, the incremental effect of education on the prevalence of dysthymia was less evident and depended on gender (Figure 1 ▶).
FIGURE 1—
Prevalence of dysthymic disorder, by years of education, race/ethnicity, and gender.
DISCUSSION
Most previous studies have suggested that African American individuals have lower rates of depression compared with White individuals. These National Health and Nutrition Examination Survey III findings indicate that prevalence of depression differs significantly by race/ethnicity but that comparative rates depend on the type of depression. African American and Mexican American individuals have higher lifetime prevalence rates of dysthymic disorder, whereas White individuals have higher lifetime prevalence rates of major depressive disorder.
What is it about being African American or Mexican American in the United States that results in chronic dysphoria? Our findings are partially explained by poverty and lack of education; however, other cultural factors may have a mediating effect. There may be subgroups of Mexican American persons (e.g., immigrants with little education who do not speak English) with a higher prevalence of dysthymic disorder. Past research has shown the importance of immigration status and generational differences (e.g., acculturation).10 We found that non–English-speaking persons had significantly lower education (P = .000). Although the Spanish-language version of the Diagnostic Interview Schedule8 was used, non–English-speaking persons may have understood questions differently, may have manifested depression differently, or may have been less willing to endorse depression. Differences by race/ethnicity in help seeking, accessing mental health services, and using psychotropic medication also may have contributed. Riolo et al.,11 in their analyses of National Health and Nutrition Examination Survey III data, found that rates of psychotropic medication treatment differed by racial/ethnic group. Although African American and Mexican American individuals have lower rates of major depressive disorder than do White individuals, they are also less likely to receive medical treatment, which may contribute to more chronic depression.11,12
Study limitations include year of data collection and consequent use of DSM-III-R diagnostic criteria. However, respondents were asked whether they had taken medication or told a provider about their symptoms; these questions were used to approximate the functional impairment criterion of Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition. In addition, accuracy of lifetime diagnoses based on self-report may be affected by recall bias (e.g., results show higher lifetime rates of major depressive disorder among those aged 20–24 vs 25–34 years). Self-report is also limited by patient insight and does not allow for examiner ability to elicit nonverbal signs of depression (e.g., psychomotor retardation). Future research is needed to consider other potentially important factors, such as unemployment, rural residence, and comorbid disorders (e.g., substance use and anxiety disorders).
Acknowledgments
We thank the Department of Psychiatry at the University of Michigan for providing financial support for this work.
We thank Kathleen Welch, Kenneth Guire, and the Center for Statistical Consultation and Research at the University of Michigan for assistance with data analysis. In addition, we thank Christen Flack for her assistance with literature review.
Human Participant Protection This research project was exempt from formal review through the institutional review board at the University of Michigan because it consisted solely of secondary data analysis of a public use database.
Peer Reviewed
Contributors S. A. Riolo originated the study and was actively involved in all aspects of its implementation, including data analysis and manuscript preparation. T. A. Nguyen assisted with data analysis, manuscript preparation, and editing of the brief. J. F. Greden assisted with study origination, data interpretation, manuscript preparation, and editing of the brief. C. A. King supervised all aspects of the study and assisted with data interpretation, manuscript preparation, and editing of the brief.
References
- 1.Somervell PD, Leaf PJ, Weissman MM, Blazer DG, Bruce ML. The prevalence of major depression in black and white adults in five United States communities. Am J Epidemiol. 1989;130:725–735. [DOI] [PubMed] [Google Scholar]
- 2.Neighbors HW, Jackson JS, Bowman PJ, Gurin G. Stress, coping, and Black mental health: preliminary findings from a national study. Prev Hum Serv. 1983; 2(3):5–29. [DOI] [PubMed] [Google Scholar]
- 3.Warheit GJ, Holzer CE, Arey SA. Race and mental illness: an epidemiologic update. J Health Soc Behav. 1975;16:243–256. [PubMed] [Google Scholar]
- 4.Kessler RC, McGonagle KA, Zhao S, et al. Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States: results from the National Comorbidity Survey. Arch Gen Psychiatry. 1994; 51:8–19. [DOI] [PubMed] [Google Scholar]
- 5.Blazer DG, Kessler RC, McGonagle KA, Swartz MS. The prevalence and distribution of major depression in a national community sample: the National Comorbidity Survey. Am J Psychiatry. 1994;151: 979–986. [DOI] [PubMed] [Google Scholar]
- 6.Kessler RC, Berglund P, Demler O, et al. The epidemiology of major depressive disorder: results from the National Comorbidity Survey Replication (NCS-R). JAMA. 2003;289:3095–3105. [DOI] [PubMed] [Google Scholar]
- 7.Weissman M, Bruce ML, Leaf PJ, Florio LP, Holzer CE. Affective disorders. In: Robins LN, ed. Psychiatric Disorders in America. New York, NY: Free Press; 1991:53–80.
- 8.Robins LN, Helzer JE, Orvaschel H, et al. The Diagnostic Interview Schedule. In: Eaton W, Kessler RC, eds. Epidemiologic Field Methods in Psychiatry: The NIMH Epidemiologic Catchment Area Program. London, England: Academic Press Inc; 1985: 143–170.
- 9.Diagnostic and Statistical Manual of Mental Disorders, Revised Third Edition. Washington, DC: American Psychiatric Association; 1987.
- 10.Hovey JD, King CA. Acculturative stress, depression, and suicidal ideation among immigrant and second-generation Latino adolescents. J Am Acad Child Adolesc Psychiatry. 1996;35:1183–1192. [DOI] [PubMed] [Google Scholar]
- 11.Riolo SA, Nguyen TA, King CA. Antidepressant medication use by age and gender: nationally representative data. Poster presented at: 50th annual meeting of the American Academy of Child & Adolescent Psychiatry; October 16, 2003; Miami, Fla.
- 12.Riolo SA, Nguyen TA, King CA. Depression prevalence and helpseeking behavior among US adolescents. Poster presented at: 14th annual Silverman Conference on “The Treatment of Depression in Real World Settings”; May 28, 2003; Ann Arbor, Mich.