Abstract
Despite the rapid growth in the number of Korean Americans (KAs) and the alarmingly high prevalence of depression in this population, relatively little is known about the impact of depression literacy on KAs’ depression. This study investigated the prevalence of depressive symptoms among a community-based sample of KA adults; the associations among socio-demographic characteristics, depression literacy, and depressive symptoms; and predictors related to depressive symptoms. A sample of 600 KAs completed depression and depression literacy scales, as well as socio-demographic questionnaires. Data analyses were conducted to assess the association between depressive symptoms, depression literacy, and the covariates. Forty-five percent of participants showed elevated depressive symptoms. Marital status, attained educational level, perceived mental health, and depression literacy were significant predictors of depressive symptoms. Given the high prevalence of depressive symptoms in this population, future research should investigate the causal relationships of various predictors of depression and depression literacy, which will facilitate the development of culturally-appropriate interventions and policies concerning mental health for KAs, and early mental health screening for them.
Keywords: Depressive Symptoms, Depression Literacy, Depression Prevalence, English Proficiency, Korean Americans
BACKGROUND
The World Health Organization (WHO) [1] reported that depression accounts for 4.3% of the global disease burden and was the third leading cause of disability worldwide, accounting for 11% globally of all years lived with disability. It further predicted that by 2030, depression would be the leading cause of disease burden globally. According to a new report from the WHO [2], depression is already the leading cause of disability and ill health worldwide; more than 300 million people are now living with depression, an increase of more than 18% between 2005 and 2015. Yet Gabriel and Violato [3] argue that depression is still not well-understood by health professionals, patients, and the general public, particularly with regards to its etiology, signs, symptoms, and treatment. Swami [4] reports that extant literature has tended to treat respondents with depression as a fairly homogenous group of people and has not fully considered individual differences in knowledge and understanding of depression as mental illness and its impact on treatment outcomes. Other depression studies have reported that negative attitudes toward professional help [12], including cultural mistrust among ethnic minority groups [13] and skepticism of psychiatry as a science [14], are significantly associated with under-utilization of health care services. Thus, even though depression is one of the most common mental disorders in the US [5], a large number of patients with it may neither recognize the illness nor the need for treatment.
Mental health literacy (MHL) recently has attracted interest from healthcare workers and researchers to explicate diagnostic differences in mental illness and its related treatment. MHL is defined as knowledge and beliefs about mental disorders that aid their recognition and management [6], and the perceived abilities necessary for maintaining mental health and preventing mental health problems [6, 7]. Therefore, MHL plays an important role in both the recognition of mental health problems, including depression, and help-seeking for those problems.
Depression literacy (DL) is a specific type of MHL, defined as the ability to recognize symptoms of depression and to make informed decisions about treatment [8]. Francis and colleagues [9] claim that depression and limited DL often co-exist, as DL among patients and the general population influences help-seeking behavior. Gazmararian et al. [10] found that persons with limited DL were nearly 3 times more likely to have depressive symptoms than were those with adequate DL. Other studies have indicated that depressed individuals with limited DL skills respond better to standard medical and depression treatments if their DL is improved [9]. Deen et al. [11] reported a DL rate of 53% among a rural American sample and recommended that MHL, including DL, could help individuals recognize, manage and prevent mental illnesses.
DL varies among different ethnic groups. Asian Americans (AAs) tend to view depressive symptoms as a sign of weakness or as a lack of discipline and willpower [15]. As a result, they tend not to accept the medical model of depression [16]. Rather, commonly shared Confucian ethics lead them to believe that self-concealment of emotional trouble is a virtue [17]. Therefore, emotional symptoms and distress are often internally tolerated, and stoic approaches to emotional distress, such as the use of willpower and the avoidance of morbid thoughts, are usually encouraged [18, 19]. Consequently, knowledge and understanding of depression as a form of mental illness are not culturally acceptable to AAs, and limited DL is inevitable.
A review of the literature indicated that there are sufficient health and MHL studies among the general United States (US) population, but we found a paucity of studies focused on DL among Korean Americans (KAs), despite the high prevalence of depression in this ethnic group. A literature review of depression prevalence among AA adults reported 35.6% experienced a high rate of depressive symptoms. The prevalence of depression among different subethnic AA groups was estimated to be as follows: 34.4% for Filipinos, 33.3% for Koreans, 20.4% for Japanese, and 15.7% for Chinese [20]. However, only two percent of AAs will mention symptoms of depression to their doctor, compared to the national average of 13% of the general population [21]. Another study found that most AAs tend to seek out support for their mental health concerns from personal networks, such as close friends, family members, and religious community members, rather than seek professional help [22]. Additionally, a study conducted in South Korea found an average of 3.4 years delay in seeking treatment after the onset of depression [23], similar to 4.0 years in the general U.S. population (24).
While the benefits of health literacy have been increasingly recognized and are widely accepted as a key predictor of health behaviors and outcomes, knowledge about mental disorders, including depression and DL among KAs, has been comparatively neglected. KAs are currently ranked as the fifth largest AA subgroup in the US, and they are projected to have continuous demographic growth over the next few decades [26]. Within the AA population, the highest suicide rates were found in both KA men and women [27]. Especially among elders (age range 65+), the suicide rate was the highest of all examined US racial and ethnic groups [28]. Therefore, it is important to study depressive symptoms and DL in this population in order to create effective mental health interventions for this underserved population.
It is important to recognize that there are many interrelated factors among DL, depressive symptoms, and socio-demographic characteristics, such as gender, education, culture, language, race, ethnicity, and socioeconomic status [25], that tend to be predictors of depressive symptoms. The current study aims to explore the significance of these factors and their influences to KAs’ overall health status. The purposes of this study are to examine: 1) the prevalence of depressive symptoms in a community-based sample of KA adults, 2) the associations among socio-demographic characteristics, depression literacy, and depressive symptoms, and 3) predictors related to depressive symptoms.
METHODS
This is an exploratory study with a goal of generating formal hypotheses and hopefully leading to intervention program based on the results. This study employed a cross-sectional design, collecting data at one time point using survey questionnaires.
Participants
Following review and approval by the Hunter College Institutional Review Board, participants aged 21 years or older who self-identified as KA and self-reported that they were able to read and write fluently in Korean or English were recruited from Korean community centers, religious organizations, and hospitals in New York City, Long Island, and Teaneck, New Jersey.
Procedures
Participants were recruited by bilingual (English and Korean) interviewers. A total of 693 participants who met the inclusion criteria participated in the self-administered paper/pen survey, which was given in either Korean or English, depending upon each participant’s preference. The survey took approximately 20–30 minutes to complete.
Before administering the survey, interviewers described the study purposes, procedures, the scope of questions that would be asked, confidentiality procedures, the right to withdraw at any time or to refuse to answer any questions, and the overall voluntary nature of participation, as well as the benefits and risks of the study. Informed consent was reviewed prior to administering the survey, and participants were offered $10 compensation for their time. Data were collected from January 2018 through May 2018.
Survey Measures
The survey questionnaires consisted of socio-demographic questionnaires, the 20-item Center for Epidemiologic Studies Depression Scale (original CES-D and CES-D-K, Korean version) to assess depressive symptoms, and a 22-item Depression Literacy Questionnaire (D-Lit) for DL.
Socio-demographic characteristics included current age, gender, level of education attained, marital status, religion, duration of residence in the US, and English proficiency. Educational attainment was classified as elementary school or less, middle school, high school, and college or higher. Marital status was classified as married, single, widowed, divorced, separated never married, and living with someone. Religion was classified as Protestant, Buddhist, Catholic, other religion, and no religion. Duration of residence in the US in years was assessed with a question about how long they had lived in the US. Self-reported English proficiency was assessed with a question, “What do you think of your English proficiency?”, which was rated as 1= “excellent,” 2= “good,” 3= “fair,” and 4= “poor.”
Subjective health variables were measured for participants’ self-perceptions of their physical and mental health, which were assessed by using ordinal variables coded as 1= “excellent,” 2= “good,” 3= “fair,” 4= “terrible.”
Both Socio-demographic questionnaire including subjective health variables were written in Korean and then translated to English by two bilingual (Korean–English) scholars and back-translated to Korean by another bilingual scholar, and then each translated item was reviewed by the research team collectively.
The Center for Epidemiologic Studies Depression Scale (CES-D) and CES-D Korean (CES-D-K) version were used to measure depressive symptoms. The CES-D is one of the most widely used self-report instruments for depression screening in the general population [30, 31]. It has 20 items, each with response options ranging from 0 to 3 where 0= “rarely or none of the time,” 1= “some or little of the time,” 2= “moderately or much of the time,” and 3= “most or almost all the time.” Four questions were reverse coded as recommended. The total composite score is calculated by summing up the individual responses. Composite scores range from 0 to 60, with higher scores indicating greater depressive symptoms. A score of 16 points or more is considered depressed [31]. Internal consistency of the CES-D was 0.85 in the general population and test–retest correlations were between 0.45 and 0.70 [31]. The CES-D-K was validated by Chon, Choi, and Yang [32], with a Cronbach’s alpha of .91. In this study, the coefficient of the combined CES-D and CES-D-K was .89.
Depression Literacy Questionnaire (D-Lit), developed by Griffiths [33], was used to assess DL after obtaining permission to use. The questionnaire consists of 22 items, each with one of three response options: “true (1),” “false (0),” or “don’t know (0).” Scoring is calculated as the sum of correct items for a maximum of 22 points. Each correct response receives one point, and “don’t know” is counted as an incorrect response [33]. The internal consistency was .70 and the test-retest reliability was r= .71 [33]. The original D-Lit questionnaire was translated to Korean by two bilingual (Korean–English) scholars and back-translated to English by another bilingual scholar, and then each translated item was reviewed by the research team collectively. There was no previous validity testing of the Korean translated D-Lit instrument available. The Cronbach’s alpha for the D-Lit was .81 in this study.
Data Analysis
Exploratory data analysis was conducted to determine the shape and distributions of variables to determine whether transformations were necessary and also to flush out implausible values. The Kolmogorov-Smirnov goodness of fit test was used to assess normality of the CES-D composite score [34]. Age was classified into the following age-groups: younger than 30, 30–49, 50–69, 70 years and older. Years in the US was categorized as shorter than 5 years, 5–10, 11–14, and 15 years or longer.
Descriptive statistics were computed for the overall sample and then stratified by high depressive score (≥16) vs. low (<16). Quantitative measures were summarized using the mean and standard deviation. Qualitative and ordinal variables were summarized using frequencies and percentages. We compared whether there were substantial differences between participants with high depressive vs. low depressive symptoms. A two-sample t-test was used to compare sociodemographic characteristics between high and low depressive score categories. A Chi-squared test was used to compare proportions between the two groups. The linear association of sociodemographic characteristics with the composite depressive score was assessed using age- and sex-adjusted Pearson correlation coefficients for continuous measures, and a Spearman rank correlation for ordinal variables. The ɳ2 (eta-squared) expressed as a percentage was used to estimate the multivariable-adjusted proportion of the total sample variance in depressive scores that is explained by each characteristic after accounting for confounding variables: sex, age, marital status, education, years lived in the US, religion, English language proficiency, perceived physical health, perceived mental health, regular medical check-ups, income, and DL. Cohen’s rule of thumb classifies an eta-squared .01 to <0.06 as small, 0.06 to <0.14 as medium, and put ≥.14 as large [35].
We used multivariable analysis of covariance (ANCOVA) to fit 3 nested linear regression models, each with the depressive symptoms composite score as the dependent variable. Model 1 included age, sex, education, marital status. Model 2 adjusted Model 1 for health-related factors: medical check-ups, perception of physical health, and perception of mental health. Model 3 adjusted Model 2 for health literacy composite score. Models were summarized using regression coefficients. Model goodness-of-fit was assessed using the coefficient of variation, R2, which measures the proportion of variation in the CES-D composite score, which is explained by the covariates in each of the three nested models. The quality of each model relative to each of the other models was assessed using the Akaike Information Criteria (AIC) [36], 37]. We tested for heterogeneity of sex and age using an interaction term in the ANCOVA model. A p-value of <0.05 was considered statistically significant. Data was analyzed using the PROC GLM and PROC GLMSELECT subroutines in the Statistical Analysis Software (SAS) version 9.4 for Windows [38]. Data and computer code are available upon reasonable request.
RESULTS
Description of Sample and Prevalence of Depressive Symptoms
Of the 693 participants, 600 (87%) had non-missing CES-D data. The 93 with missing data were excluded. Also excluded from regression models were 45 participants with missing socio-demographic variables. Sample characteristics are shown on Table 1.
Table 1.
Description of Sample Characteristics and Prevalence of Depressive Symptoms
| All | Depressive Symptoms-CES-D Scale | ||||
|---|---|---|---|---|---|
| Characteristics | N=600 | Low <16, n=328 | High ≥16, n=272 | p-value* | |
| CES-D composite score | 15.7(9.3) | 8.9(4.3) | 23.9(6.9) | <.0001 | |
| Gender | |||||
| Female | 339(56.7) | 181(55.4) | 158(58.3) | 0.507 | |
| Male | 259(43.3) | 146(44.6) | 113(41.7) | ||
| Age (years) | 52.7(18.9) | 52.6(17.9) | 52.7(20.0) | 0.937 | |
| Age group | |||||
| <30 | 99(16.5) | 46(14.0) | 53(19.5) | 0.073 | |
| 30-<50 | 183(30.5) | 108(32.9) | 75(27.6) | ||
| 50-<70 | 158(26.3) | 94(28.7) | 64(23.5) | ||
| ≥70 | 160(26.7) | 80(24.4) | 80(29.4) | ||
| Marital Status | |||||
| Married/ Cohabiting | 383(64.4) | 236(72.6) | 147(54.4) | <0.001 | |
| Widowed | 43(7.2) | 18(5.5) | 25(9.3) | ||
| Divorced/Separated | 39(6.6) | 13(4.0) | 26(9.6) | ||
| Never Married | 130(21.9) | 58(17.9) | 72(26.7) | ||
| Education | |||||
| Elementary and less | 13(2.2) | 1(0.3) | 12(4.4) | <0.001 | |
| Middle School | 27(4.5) | 7(2.1) | 20(7.4) | ||
| High School | 148(24.8) | 71(21.7) | 77(28.5) | ||
| 4-Year College + | 409(68.5) | 248(75.8) | 161(59.6) | ||
| Years in the US (years) | 22.0(12.1) | 22.0(12.7) | 22.1(11.4) | 0.962 | |
| Years in the US | |||||
| <5 years | 38(6.5) | 23(7.1) | 15(5.7) | 0.703 | |
| 5-<10 years | 62(10.6) | 37(11.5) | 25(9.4) | ||
| 10-<15 years | 70(11.9) | 39(12.1) | 31(11.7) | ||
| ≥15 years | 417(71.0) | 223(69.3) | 194(73.2) | ||
| Religion | |||||
| No Religion | 42(7.1) | 21(6.5) | 21(7.8) | 0.034 | |
| Christian | 443(74.5) | 253(78.1) | 190(70.1) | ||
| Catholic | 77(12.9) | 32(9.9) | 45(16.6) | ||
| Buddhist | 23(3.9) | 15(4.6) | 8(3.0) | ||
| Other | 10(1.7) | 3(0.9) | 7(2.6) | ||
| English Language Proficiency | |||||
| Excellent | 74(12.8) | 40(12.7) | 34(12.8) | 0.006 | |
| Good | 194(33.5) | 121(38.5) | 73(27.6) | ||
| Fair | 280(48.4) | 143(45.5) | 137(51.7) | ||
| Terrible | 31(5.4) | 10(3.2) | 21(7.9) | ||
| Perceived Physical Health | |||||
| Excellent | 79(13.6) | 48(15.3) | 31(11.7) | 0.<0.001 | |
| Good | 321(55.4) | 202(64.3) | 119(44.9) | ||
| Fair | 147(25.4) | 54(17.2) | 93 (35.1) | ||
| Terrible | 32(5.5) | 10(3.2) | 22(8.3) | ||
| Perceived Mental Health | |||||
| Excellent | 117(20.2) | 84(26.7) | 33(12.5) | <0.001 | |
| Good | 322(55.5) | 197(62.5) | 125(41.2) | ||
| Fair | 124(21.4) | 30(9.5) | 94(35.5) | ||
| Terrible | 17(2.9) | 4(1.3) | 13(4.9) | ||
| Regular Medical Check-up | |||||
| Yes | 364(63.2) | 207(65.9) | 157(59.9) | 0.141 | |
| No | 212(36.8) | 107(34.1) | 105(40.1) | ||
| Income | |||||
| <$5,000 | 52(10.2) | 28(10.6) | 24(9.7) | 0.030 | |
| $5,000-$9,999 | 52(10.2) | 20(7.6) | 32(13.0) | ||
| $10,000-$19,999 | 60(11.7) | 29(11.0) | 31(12.6) | ||
| $20,000-$29,999 | 65(12.7) | 28(10.6) | 37(15.0) | ||
| $30,000-$39,999 | 54(10.6) | 27(10.2) | 27(10.9) | ||
| $40,000–49,999 | 55(10.8) | 25(9.5) | 30(12.2) | ||
| $50,000+ | 173(33.9) | 107(40.5) | 66(26.7) | ||
| Depression literacy, Mean/SD | 9.4(4.4) | 9.6(4.6) | 9.1(4.1) | 0.147 | |
p value based on t-test for continuous characteristics, and Chi-squared test for categorical variables.
Mean age was 52.7 years; 16.5% were younger than 30 years of age, and the rest were evenly distributed in age-groups older than 30 years. More than one half, 56.7%, were female. Sixty-four percent were either married or cohabiting while more than two-thirds had at least a bachelor’s degree. A large majority, 71%, had lived in the US for longer than 15 years and an equal proportion reported being of Christian faith. Two-thirds reported having good to excellent physical health, and 3 out of 4 perceived their mental health to be good/excellent. More than 1 in 3 reported household income greater than $50,000 annually. Fewer than half (46.3%) rated their proficiency in English language as good to excellent. 45.3% (n=272) had CES-D scores of 16 or greater, indicative of clinical depression. Distributions of high vs. low CES-D groups were similar with respect to age, sex, years in the US, and having regular medical check-ups. Participants with high CES-D scores were more likely to be have been widowed, divorced/separated, or never married (p<0.001), and were more likely to have lower levels of education (p<0.001). They also were less likely to rate their proficiency in English language as good/excellent (p=0.006), rate their physical health as good/excellent (p<0.001), and rate their mental health as fair/excellent (p<0.001) (Table 1).
Associations among Demographics, Depression Literacy and Depressive Symptoms
Table 2 shows age- and sex-adjusted measures of association of depressive symptoms scores with demographic characteristics. Higher levels of educational attainment (r=−0.26; p<0.0001), higher income (r=−0.15; p<0.001), higher English language proficiency (r=−0.12; p<0.05), higher perceived physical health (r=−0.15; p<0.0001), higher perceived mental health (r=−0.38; p<0.0001), and higher DL (r=−0.13; p<0.01) were negatively linearly associated with depression symptoms composite scores. Regular medical check-ups (r=0.09; p<0.05) was positively linearly associated with depression symptoms. There was no linear association with age and years living the US (Table2).
Table 2.
Association of Depressive Symptoms Score with Demographic Characteristics
| Characteristics | Correlation coefficient, r | Semipartial, ɳ2 (%) |
|---|---|---|
| Sex | - | 0.61 |
| Age | 0.05 | 0.82 |
| Marital Status | - | 5.89 |
| Education | −0.26**** | 8.24 |
| Years in the US | 0.06 | 0.42 |
| Religion | - | 0.31 |
| English Language Proficiency | −0.12 * | 0.61 |
| Perceived Physical Health | −0.28**** | 4.20 |
| Perceived Mental Health | −0.38**** | 6.22 |
| Regular Medical Check-up | 0.09* | 0.44 |
| Income | −0.15*** | 1.22 |
| Depression literacy | −0.13** | 0.40 |
p < .05
p < .01
p < .001
p<0.0001
Figure 1 and Table 2 shows covariate-specific, semi-partial ɳ2 (eta-squared), representing the percentage of variation in depressive score explained by each characteristic. Education, perception of mental health and marital status had the largest eta-square, classified as medium using Cohen’s criteria. Perception of physical health and income explained proportions classified as small.
Figure 1.
ɳ2 (eta-squared) as a percentage of proportion of variation in depressive score explained by each characteristic
Predictors Related to Depressive Symptoms
Table 3 shows linear regression coefficients for three nested models of depressive symptoms as the dependent/response variable. Model 1 was minimally adjusted for age, sex, educational attainment, and marital status. Model 1 accounted for 13.0% of the variation in depressive scores. Model 2 adjusted Model 1 for regular medical check-ups, perceived physical health, and perceived mental health. It explained 24.2% of the variation. Model 3 adjusted Model 2 for DL. It explained 26.9% of the variation (Figure 2). AICs for the three models show that there was improvement in model fit between Model 1 vs. Model 2 (AIC 3029.2 vs. 2965.9) and between Model 2 vs. Model 3 (AIC: 2965.9 vs. 2743.4). Contributing substantially to Models 1, 2, and 3 were the following predictors: educational attainment, marital status, perceived mental health, and DL. As shown in Figure 1, each of these four factors individually accounted for between 4% and 9% of the variability in CES-D scores. A closer inspection of the eta-squared values reveals that participants who graduated from elementary school and middle school and were never married reported higher depressive symptoms. Additionally, those who have poor mental health reported higher depressive symptoms. Lastly, participants having a lower DL reported higher depressive symptoms.
Table 3.
Regression Coefficients for Three Nested Models of Depressive Symptoms
| Model 1a | Model 2b | Model 3c | ||
|---|---|---|---|---|
| Age | 0.02 | −0.01 | −0.02 | |
| Gender | Male | −0.16 | −0.32 | −0.59 |
| Female | ref | ref | ref | |
| Education | ≤Elementary | 13.98**** | 11.47*** | 11.23**** |
| Middle School | 8.70**** | 6.70*** | 6.13*** | |
| High School | 2.99*** | 2.34** | 2.07* | |
| 4-Year College + | ref | ref | ref | |
| Marital status | Divorced/Separated | 3.57* | 1.81 | 1.93 |
| Never Married | 3.06** | 2.22* | 2.24* | |
| Widowed | 2.00 | 1.22 | 1.74 | |
| Married/ Cohabiting | ref | ref | ref | |
| Medical Check-ups | No | 0.78 | 1.16 | |
| Yes | ref | ref | ||
| Physical health | Good | −1.01 | −0.69 | |
| Fair | 0.95 | 1.21 | ||
| Terrible | 0.99 | 1.27 | ||
| Excellent | ref | ref | ||
| Mental health | Good | 2.85 | 2.89** | |
| Fair | 7.68**** | 7.95**** | ||
| Terrible | 10.49**** | 11.45**** | ||
| Excellent | ref | ref | ||
| Depression Literacy | −0.21* | |||
| Model Goodness of fit (Model R2) | 13.03% | 24.20% | 26.86% | |
| Akaike Information Criteria (AIC) | 3029.2 | 2965.9 | 2743.4 | |
p < .05
p < .01
p < .001
p<0.0001
Model 1, adjusted for age, sex, education attainment, marital status, 13.0% of variance explained by Model 1.
Model 2, adjusted Model 1 for regular medical checkups, perceived physical health, and perceived mental health, 24.2% of variance explained by Model 2.
Model 3, adjusted Model 2 for depression literacy, 26.9% of variance explained by Model 2.
Figure 2.
Proportion of variation in CES-D explained by three nested models
Model 1*, adjusted for Age, Sex, Education attainment, marital status, Model AIC=3029.2
Model 2**, adjusted Model 1 for Regular Medical Check-ups, Perceived Physical and Mental, Model, AIC=2965.9
Model 3***, adjusted Model 2 for depression literacy, Model AIC=2743.4
DISCUSSION
This study examined the prevalence of depressive symptoms among KAs, the association of symptoms with demographic characteristics and DL, and predictors of depressive symptoms. Findings indicate that nearly half (45.3%) of the participants had depressive symptoms at or above the cutoff score of 16 on the CES-D [31]; this prevalence is much higher than the 23.0% found in our previous study of KA immigrants [39] and is more than twice that of American adults (20.1%) [40]. Additionally, this study suggests that KAs are four times more likely to exhibit depressive symptoms than are native Koreans in South Korea, as a 2009 cross-sectional Korean Community Health Survey of 229,595 Korean adults, ages 19 and older, found that 11% had depressive symptoms (7.8% for men, 14.0% for women) [41].
The present study found that KA participants who reported higher educational attainment, higher income levels, higher English language proficiency, higher perceived mental and physical health, and higher DL had a lower level of depressive symptoms. However, gender, age, marital status, duration in the US, and religion had no association with depressive symptoms. These findings are partially supported by Gülfizar’s [42] study, which found that being female, a widower or single, having a low education level, and comorbid physical illness were associated with depression among the elderly. Furthermore, another study, which examined the prevalence of depression among older KAs, provided additional support for the finding that English language proficiency and DL are both associated with depression [43].
In addition, the study found no significant association with depressive symptoms in either sex, even though the relationship between sex and depression are well-established in the scientific literature--i.e., studies of the U.S. general population have found greater prevalence of depression in women than in men [44, 45, 46]. In contrast, Lisznyai et al. [47] reported that male Hungarians had higher scores of depressive symptoms than their female counterparts, and Cho et al. [48] reported that KA men (30.5%) reported a significantly higher age-adjusted prevalence of depressive symptoms than did KA women (19.5 %).
In this study, a lower level of education attainment was associated with a higher burden of depressive symptoms, which is supported by findings in other studies [39, 49, 50, 43]. Low educational attainment is often accompanied by low levels of English proficiency and in low DL, which can lead to a higher propensity for depressive symptoms. Participants’ annual household income was one of the socio-demographic factors negatively correlated with depressive symptoms. Cho and colleagues [48] and Bernstein and colleagues [39] strongly support that household income is indirectly associated with depression among KAs. This study also found a significant association between perceived mental health and depressive symptoms. In contrast, Fleishman and Zuvekas [51] and Cho and colleagues [48] reported that perceived mental health did not directly correspond with measured (or diagnosed) mental disorders, such as depression. Regular medical check-ups had a positive association with depressive symptoms, potentially because they likely increase the propensity for depression detection.
Self-reported English proficiency was negatively correlated with depression, a finding consistent with Bernstein et al.’s [39] study of KA immigrants, and with Baird et al.’s [43] study of older KAs. Furthermore, a study by Kim et al. [52] of older Mexican American adults reported that limited English proficiency was a risk factor for not only greater depressive symptomatology, but also accelerated the progression of depressive symptoms over time. Jang and colleagues [53] reported that limited English proficiency was a critical source of health vulnerability and that systematic efforts should be made in health-related research and prevention and intervention programs for older Korean Americans.
Higher levels of DL among KA respondents revealed lower burden of depressive symptoms. This result is consistent with previous studies that reported that depression and limited DL often co-exist [9], and that persons with limited DL were more likely to have depressive symptoms relative to those with higher DL [54]. Jang et al.’s [55] study of elderly Koreans reported similar findings. Studies consistently report that low health literacy impedes the ability of AAs to obtain, process, understand, and communicate health information, which results in limited access to early detection and treatment [29, 56, 57]. The high prevalence of depressive symptoms found in the present study and the higher levels of depressive symptoms in participants with poorer knowledge of depression underscores the need for educational interventions to improve the MHL of KA adults.
Multivariable-adjusted linear regression models identified education, marital status, perceived mental health, and DL to be associated with depressive symptoms. Interestingly, KA adults who lived with a spouse or someone else reported lower levels of depressive symptoms compared to participants who were divorced/separated or never married. Previous studies have consistently revealed that within the KA population, married individuals exhibit better mental health compared to single, separated, or divorced individuals [41, 58]. Our finding indicated that married individuals, compared to their single counterparts, have better emotional and social support as well as meaning and purpose in life, attributes which promote mental health and reduce the likelihood of depression.
Another important finding was that KA participants who reported poor mental health and a lower level of DL were more likely to experience an increase in depressive symptoms. Gazmararian and colleagues [10] reported a similar finding that people with limited DL showed depressive symptoms three times higher than those with a higher level of DL. Thus, DL is a key factor affecting the risk of developing depression in the KA population. A limited DL is deeply coupled with KA cultural values that stigmatize depressive symptoms as a sign of weakness, and a lack of discipline and willpower [18,19]. Consequently, many KAs do not seek appropriate information about depression and mental health services and treatment, placing them at a greater risk for compromised mental health. Numerous studies suggest that DL can help individuals recognize, manage, and prevent mental illness [11,12,13]. Thus, there is an urgent need to develop culturally-tailored and linguistically-attuned educational programs that address DL in a targeted manner, especially to KAs who live alone, perceive their mental health to be poor, and have a lower level of educational attainment.
LIMITATIONS AND STRENGTHS
Several limitations of the present study should be acknowledged. The use of a convenience sampling strategy may bring a possibility of selection bias, thereby possibly impacting the levels of depressive symptoms and DL found. In addition, the current study sample was mainly recruited from KA community centers, religious organizations, and an Asian medical program in a hospital setting in the New York City metropolitan area. Thus, findings cannot be generalized to other KA populations in different settings (e.g., other large cities where significant numbers of KAs reside, such as Los Angeles, San Francisco, Chicago, or rural areas without a KA community). The cross-sectional design of the study limits any causal inferences or longitudinal conclusions between socio-demographic characteristics and depressive symptoms, and between socio-demographic characteristics and DL. In addition, data was collected on a self-report basis with no external validations. Still, this study is the first to examine the direct and indirect effects of limited English proficiency on depressive symptoms, and provides insight into the unique contribution of KAs’ socio-demographic characteristics, depressive symptoms, and DL.
Despite these limitations, this study has several strengths: the large sample size (more than 600 participants), and the multiple forms of data analyses used to ascertain the associations, including adjusted correlations, ANCOVA, and eta-squared for each predictor. We also benchmarked the eta-squared against Cohen’s criteria.
New Contributions to the Literature and Directions for Future Studies
KAs’ high prevalence of depression is well known, but relatively little is known about the impact of DL on their depression. Given the study findings of a high prevalence of depressive symptoms, and that those with higher levels of depressive symptoms exhibited lower DL, there is a clear need for educational interventions that improve DL among KA adults. Future studies may examine how an improvement in English proficiency predicts changing levels of perceived mental health and depressive symptoms. Furthermore, future research needs to investigate the causal relationships of various predictors on depressive symptoms, as this would contribute to facilitating early mental health screening and to developing culturally appropriate interventions and policies concerning KA mental health.
Acknowledgements
The research team would like to acknowledge the participants involved in this study who shared their valuable information with us. Research reported in this publication was supported by the National Center for Advancing Translational Sciences of the National Institutes of Health under Award Number UL1TR002384. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Footnotes
Compliance with Ethical Standards
Conflict of interest No competing financial interests exist.
REFERENCES
- 1.World Health Organization (WHO): Mental health action plan 2013 – 2020. 2013: WHO Library Cataloguing-in-Publication Data. WHO Press, World Health Organization, Geneva, Switzerland. [Google Scholar]
- 2.World Health Organization (WHO): “Depression: let’s talk.” 2017: WHO Press, World Health Organization, Geneva, Switzerland. [Google Scholar]
- 3.Gabriel A, Violato C. Depression literacy among patients and public: A literature review. Primary Psychiatry. 2010;17: 5–64. [Google Scholar]
- 4.Swami V. Mental health literacy of depression: Gender differences and attitudinal antecedents in a representative British sample. PLOS ONE. 2012;7(11): e49779. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Baumeister H, Harter M. Prevalence of mental disorders based on general population surveys. Social Psychiatry and Psychiatric Epidemiology. 2007;42(7): 537–46. [DOI] [PubMed] [Google Scholar]
- 6.Jorm AF. Mental health literacy: Empowering the community to take action for better mental health. American Psychologist. 2012; 67(3): 231–243. 10.1037/a0025957 [DOI] [PubMed] [Google Scholar]
- 7.Kim EJ, Yu JH, Kim EY. Pathways linking mental health literacy to professional help-seeking intentions in Korean college students. J Psychiatric Mental Health Nursing. 2020. Jan 18. doi: 10.1111/jpm.12593. [DOI] [PubMed] [Google Scholar]
- 8.Wang J, Adair C, Fick G, Lai D, Evans B, Perry BW, Jorm A. Depression literacy in Alberta: findings from a general population sample. The Canadian Journal of Psychiatry. 2007;52(7): 442–9. [DOI] [PubMed] [Google Scholar]
- 9.Francis L, Weiss BD, Senf JH, Heist K, Hargraves R. Does literacy education improve symptoms of depression and self-efficacy in individuals with low literacy and depressive symptoms? A preliminary investigation. Journal of American Board Family Medicine. 2007;20: 23–7. [DOI] [PubMed] [Google Scholar]
- 10.Gazmararian J, Baker D, Parker R, Blazer DG. A multivariate analysis of factors associated with depression: eEvaluating the role of health literacy as a potential contributor. Archives of Internal Medicine. 2000;160: 3307–14. [DOI] [PubMed] [Google Scholar]
- 11.Deen TL, Bridges AJ, McGahan TC, Andrews AR. Cognitive appraisals of specialty mental health services and their relation to mental health service utilization in the rural population. The Journal of rural health: official journal of the American Rural Health Association and the National Rural Health Care Association, 2011;28(2): 142–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Van Voorhees BW, Fogel J, Houston TK, Cooper LA, Wang NY, Ford D. Attitudes and illness factors associated with low perceived need for depression treatment among young adults. Social Psychiatry & Psychiatric Epidemiology. 2006;41: 746–754. [DOI] [PubMed] [Google Scholar]
- 13.Soorkia R, Snelgar R, Swami V. Factors influencing attitudes towards seeking professional psychological help among South Asian students in Britain. Mental Health, Religion & Culture. 2011;14: 613–623. [Google Scholar]
- 14.Swami V, Furnham A. Preliminary investigation of the psychometric properties of the Psychiatric Skepticism Scale. Scandinavian Journal of Psychology. 2011;52: 399–403. [DOI] [PubMed] [Google Scholar]
- 15.Leong FTL, Lau ASL. Barriers to providing effective mental health services to Asian Americans. Mental Health Services Research. 2001;3(4): 201–214. [DOI] [PubMed] [Google Scholar]
- 16.Karasz A. Cultural differences in conceptual models of depression. Social Science and Medicine. 2005;60(7): 1625–35. [DOI] [PubMed] [Google Scholar]
- 17.Yi SH, Tidwell R. Adult Korean Americans: Their Attitudes Toward Seeking Professional Counseling Services. Community Mental Health Journal. 2005; 41: 571–80. [DOI] [PubMed] [Google Scholar]
- 18.Lin KM, Cheung F. Mental health issues for Asian Americans. Psychiatric Service. 1999; 50(6): 774–80. [DOI] [PubMed] [Google Scholar]
- 19.Park SY, Bernstein KS. Depression and Korean American immigrants. Archives of Psychiatric Nursing. 2008;22(1): 12–19. [DOI] [PubMed] [Google Scholar]
- 20.Kim HJ, Park E, Storr CL, Tran K, Juon HS. Depression among Asian-American adults in the community: Systematic review and meta-analysis. PLOS ONE. 2015;10(6); e0127760. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Office of the Surgeon General (US); Center for Mental Health Services (US); National Institute of Mental Health (US). Mental Health: Culture, Race, and Ethnicity: A Supplement to Mental Health: A Report of the surgeon general.: Substance Abuse and Mental Health Services Administration (US), Rockville (MD). 2001. Aug. Chapter 5. [PubMed] [Google Scholar]
- 22.Spencer MS, Chen J, Gee GC, Fabian CG, Takeuchi DT. Discrimination and mental health–related service use in a national study of Asian Americans. American Journal of Public Health. 2010;100 (12): 2410–2417. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Ki M, Paik JW, Choi KS, Ryu SH, Han C, Lee K, Ham BJ, Chang HS, Won ES, Jun TY, … Lee MS. Delays in depression treatment among Korean population. Asia-Pacific Psychiatry. 2014;6(4): 414–24. [DOI] [PubMed] [Google Scholar]
- 24.Wang PS, Angermeyer M, Borges G, et al. Delay and failure in treatment seeking after first onset of mental disorders in the World Health Organization’s World Mental Health Survey Initiative. World Psychiatry. 2007;6(3): 177–185. [PMC free article] [PubMed] [Google Scholar]
- 25.Shern D, Steverman S, Ahmed E, Shea P. Situational Analysis: Issues of Relevance in Designing a National Strategy to Promote Mental, Emotional, and Behavioral Health and to Prevent/Reduce Mental Illlness and Substance Use Disorders. Alexandria, VA: 2011: National Association of State Mental Health Program Directors. [Google Scholar]
- 26.US Census Bureau. 2010 U.S. Census Bureau: Census American fact finder. 2011: US Census Bureau, Washington, DC. [Google Scholar]
- 27.Kuroki Y1.Comparison of Suicide Rates Among Asian Americans in 2000 and 2010: Omega (Westport). 2018;77(4): 404–411. doi: 10.1177/0030222816678425. [DOI] [PubMed] [Google Scholar]
- 28.Kung A, Hastings KG, Kapphahn KIE, Wang J … Cross-national comparisons of increasing suicidal mortality rates for Koreans in the Republic of Korea and Korean Americans in the USA, 2003–2012. 10.1017/S2045796016000792. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Khuu BP, Lee HY, Zhou AQ, Shin J, Lee RM. Healthcare providers’ perspectives on parental health literacy and child health outcomes among Southeast Asian American immigrants and refugees. Children and Youth Services Review. Elsevier; 2016;67(C): 220–229. [Google Scholar]
- 30.Kazarian SS. Validation of the Armenian center for epidemiological studies depression scale (CES-D) among ethnic Armenians in Lebanon. International Journal of Social Psychiatry. 2009;55: 442–448. [DOI] [PubMed] [Google Scholar]
- 31.Radloff LS. The CES-D scale: Aa self-report depression scale for research in the general population. Applied Psychological Measurement. 1977;1(3): 385–401. [Google Scholar]
- 32.Chon KK, Choi SC, Yang BC. Integrated Adaptation of CES-D in Korea. Korean Journal of Health Psychology. 2001;6: 59–76. [Google Scholar]
- 33.Griffiths KM, Christensen H, Jorm AF, Evans K, Groves C. Effect of web-based depression literacy and cognitive-behavioral therapy interventions on stigmatizing attitudes to depression: Randomized controlled trial. British Journal of Psychiatry. 2004;185: 342–349. [DOI] [PubMed] [Google Scholar]
- 34.Massey FJ. “The Kolmogorov-Smirnov Test for Goodness of Fit.” Journal of the American Statistical Association. 1951; 46(253): 68–78. [Google Scholar]
- 35.Cohen J. Statistical Power Analysis for the Behavioral Sciences (2nd edition). 1988. Hillsdale, New Jersey: Lawrence Erlbaum Associates, Publishers. [Google Scholar]
- 36.Aho K, Derryberry D, Peterson T. Model selection for ecologists: the worldviews of AIC and BIC. Ecology. 2014;95 (3): 631–636, doi: 10.1890/13-1452.1, PMID 24804445. [DOI] [PubMed] [Google Scholar]
- 37.Kotz S, Johnson NL. (Eds) Breakthroughs in Statistics: Methodology and Distribution. Springer-Verlag. 1992; 2: 610–624. [Google Scholar]
- 38.SAS Institute Inc. Cary, NC, USA. [Google Scholar]
- 39.Bernstein KS, Park S-Y, Shin J, Cho S, Park Y. Acculturation, discrimination and depressive symptoms among Korean immigrants in New York City. Community Mental Health Journal. 2011;47(1): 24–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Oh DH, Kim SA, Lee HY, Seo JY, Choi B-Y, Nam JH. Prevalence and Correlates of Depressive Symptoms in Korean Adults: Results of a 2009 Korean Community Health Survey. J Korean Med Sci 2013;28: 128–135. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Jang SN., Kawachi I, Chang J, Boo K, Shin HG, Lee H, Cho SI. Marital status, gender, and depression: analysis of the baseline survey of the Korean longitudinal study of Ageing. Social Science & Medicine, 2009; 69 (11): 1608–1615. [DOI] [PubMed] [Google Scholar]
- 42.Gülfizar Sözeri-Varma. Depression in the Elderly: Clinical Features and Risk Factors. Aging and Disease. 2012;3(6): 465–471. [PMC free article] [PubMed] [Google Scholar]
- 43.Baird B, Oh KM, Douglas C, Weinstein AA. Health literacy, depression literacy, and depression among older Korean Americans. Journal of Health Communication. 2019; 24(5): 525–535. [DOI] [PubMed] [Google Scholar]
- 44.Smith D, Blackwood D. Depression in young adults. Advances in Psychiatric Treatment. 2004;10(1): 4–12. [Google Scholar]
- 45.Kessler RC, Chiu WT, Demler O, Merikangas KR, Walters EE. Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry. 2005;62(6): 617–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Brody DJ, Pratt LA, Hughes JP. Prevalence of depression among adults aged 20 and over: United States, 2013–2016. NCHS data brief. 2018;(303):1–8. [PubMed] [Google Scholar]
- 47.Lisznyai S, Vida K, Németh M, Benczúr Z. Risk factors for depression in the emerging adulthood. The European Journal of Counselling Psychology. 2014;3(1): 54–68. [Google Scholar]
- 48.Cho S, Park SY, Bernstein K, Roh S, & Jeon GS. Socio-demographic and Health Behavioral Correlates of Depressive Symptoms among Korean Americans. Community Mental Health Journal. 2015;51(4): 414–423. DOI 10.1007/s10597-014-9802-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Akhtar-Danesh N, Landeen J. Relation between depression and sociodemographic factors. International Journal of Mental Health Systems. 2007;1: 4. DOI: 10.1186/1752-4458-1-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Aczon-Armstrong M, Inouye J, Reyes-Salvail F. Depression and chronic illness: Asian/Pacific Islander adults in Hawaii. Issues in Mental Health Nursing. 2013;34:169–179. [DOI] [PubMed] [Google Scholar]
- 51.Fleishman JA, Zuvekas SH. Global self-rated mental health: Associations with other mental health measures and with role functioning. Medical Care. 2007;45(7): 602–609. [DOI] [PubMed] [Google Scholar]
- 52.Kim G, Kim M, Park S, Jimenez DE, Chiriboga DA. Limited English proficiency and trajectories of depressive symptoms among Mexican American elders. The Gerontologist. 2018;00: 1–9. doi: 10.1093/geront/gny032. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Jang Y, Yoon H, Park NS, Chiriboga DA. Health vulnerability of immigrants with limited English proficiency: A Study of Older Korean Americans. Journal of the American Geriatrics Society. 2016;(64)7C1: 1387–1521. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54.Gazmararian J, Baker D, Parker R, Blazer DG. A multivariate analysis of factors associated with depression: evaluating the role of health literacy as a potential contributor. Arch Intern Med. 2000;160: 3307–14. [DOI] [PubMed] [Google Scholar]
- 55.Jang Y, Gum AM, Chiriboga DA. Knowledge of depression among Korean American older adults. Journal of Applied Gerontology. 2010;(30)5: 655–665. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56.Choi SE, Liu M, Palaniappan LP, Wang EJ, Wong ND. Gender and ethnic differences in the prevalence of type 2 diabetes among Asian subgroups in California. Journal of Diabetes and Its Complications. 2013;27: 429–435. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 57.Sentell T, Baker KK, Onaka A, Braun K. Low health literacy and poor health status in Asian Americans and Pacific Islanders in Hawaii. Journal of Health Communication. 2011; 16:279–294. [DOI] [PubMed] [Google Scholar]
- 58.Bernstein k Park SY, Nokes KM Resilience and Depressive Symptoms among Korean Americans with History of Traumatic Life Experience. Community Mental Health, 2017; 53 (7): 793–801. [DOI] [PubMed] [Google Scholar]