Self-Stigma and Its Correlates Among Outpatients With Depressive Disorders

There is overwhelming evidence that individuals with depression are being seriously undertreated in both Taiwan (1) and the United States (2). Stigma associated with mental illnesses has been found to be one of the barriers to sustaining treatment for depression, suggesting the need for further study of stigma with the aim of improving treatment adherence among patients who have depression (3).

Self-stigma has been found to be widespread among depressed patients (4). However, studies of self-stigma from the perspectives of depressed patients have been few, and the results of this limited number of studies on factors associated with stigma are still controversial (4,5,6). A previous study showed that Taiwanese patients with major depression had significantly greater self-stigma than patients with nondepressive minor mental disorders (7). The Project on Stigma in Taiwanese Depressives (PSTD) was started in 2003 with the aim of exploring levels of self-stigma in a group of patients with diagnoses of depressive disorders and to further examine the relationships between self-stigma and other factors, such as severity of depressive symptoms, sociodemographic characteristics, and course of the illness.

Methods

A total of 247 patients with diagnoses of depressive disorders were recruited from outpatient psychiatric settings of three general hospitals in southern Taiwan. Ninety-three patients (40 percent) were men; 163 (66 percent) had major depressive disorders, 48 (19 percent) had dysthymic disorders, and 36 (15 percent) had depressive disorder not otherwise specified. The patients' mean±SD age was 43.9±14.3 years (range, 14 to 87 years), the mean duration of their education was 11.7±3.9 years (range, zero to 19 years), and the mean duration of their depression was 56.3± 73.9 months (range, one to 480 months). Thirty-nine patients (16 percent) had previously been hospitalized for depression, 118 (48 percent) had depressed relatives or friends, and 152 (62 percent) had low socioeconomic status.

We used the Taiwanese version of the Self-Stigma Assessment Scale (SSAS), an eight-item 5-point scale that has undergone a standardized two-way translation, to assess participants' attitudes toward their own mental illnesses, including whether they agreed that, because of mental illnesses, they were morally weak, unable to care for themselves, unable to handle responsibility, dangerous, and unworthy of respect (8). Possible total scores on the SSAS range from 8 to 40, with higher scores indicating greater self-stigma toward mental illnesses. We used the Center for Epidemiologic Studies' Depression Scale (CES-D), which is a 20-item self-administered 4-point scale, to assess participants' frequency of depressive symptoms in the preceding week (9). Possible total scores on the CES-D range from 0 to 60, with higher scores indicating more severe depression. In this study, Cronbach's alpha for the SSAS and the CES-D was .80 and .93, respectively. We also used the Social Status Rating Scale (SSRS) to categorize patients' social status into five classes, according to their employment and educational levels (10). For statistical purposes, classes 1 through 4 were labeled as "high social status," and class 5 was labeled as "low social status." We also evaluated participants' frequency of receiving knowledge about depression from any source.

The study was conducted with the approval of the institutional review board of Kaohsiung Medical University. All participants signed a statement of informed consent agreeing to interviews. Two research assistants assisted the patients in filling in questionnaires to collect information. For statistical purposes, a patient was considered to have a high level of self-stigma if the SSAS score was at least 24, which is half the total possible score on this scale. Patients who had scores of less than 24 were considered to have low levels of self-stigma. The proportions of patients with high and low levels of self-stigma were then calculated, and the predictive potentials of variables for self-stigma were examined with use of a logistic regression model.

Results

The mean±SD depression score on the CES-D for the 247 study participants was 26.1±14.7 (range, 0 to 60). The mean self-stigma score on the SSAS was 19.5±5.2 (range, 8 to 32). A total of 62 patients (25 percent) had high levels of self-stigma as defined above, with a mean score of self-stigma of 26.3±2.5 (range, 24 to 32).

The associations between self-stigma and clinical and sociodemographic characteristics were analyzed with use of a logistic regression model. As shown in Table 1, the model that included the full set of client characteristics was statistically significant (-2 log likelihood= 247.245, p<.001). Patients who had more severe depression (odds ratio [OR]=1.04, 95 percent confidence interval [CI]=1.02 to 1.07) and less education (OR=.89, CI=.81 to .98) had higher levels of self-stigma. Sex, age, duration of illness, previous admission for depression, having depressed relatives or friends, frequency of receiving knowledge about depression, and social status were not related to self-stigma.

Discussion and conclusions

In this study, more severe depression was associated with greater self-stigma, which was congruent with the results of previous studies (5,6). Pyne and colleagues (6) suggested that patients with more severe depression might in fact be socially isolated by others. Consequently, perceived higher levels of stigma are based on accurate perceptions of stigmatizing events and not due to cognitive distortions (6). We did not find an association between stigma and sex or age, although the study by Roeloffs and colleagues (4) found that older patients and female patients had higher levels of stigma that affected employment. Although previous studies for the association between stigma and employment yielded mixed conclusions (4,6), our study found that social status, which was measured according to employment and educational levels, was not associated with self-stigma. We found that the frequency of receiving knowledge about depression was not associated with self-stigma. The study by Pyne and colleagues (6) yielded the same results, indicating that educational intervention alone may not decrease stigma.

Our finding that previous hospitalization for depression and duration of illness were not related to self-stigma might indicate that patients with depression do not experience reduced self-stigma about their illness as a result of contact with medical staff in the course of episodes of depression. However, we did not examine whether mental health caregivers might be one source of stigmatizing experiences for patients with depression.

The heterogeneity of research designs and measures limits the possibility of comparing the prevalence of self-stigma among depressed patients with that of persons who have other mental illnesses. However, in our study a quarter of the depressed outpatients had high levels of self-stigma. Because patients with depression might not reduce their self-stigma during the course of their illness, clinicians must take depressed patients' viewpoints and their cultural context into consideration and try to explain the etiology and treatment plans in ways that will not exacerbate their self-stigma.

One limitation of our study is that assessment of self-stigma was restricted to participants' self-reporting on the SSAS. Another limitation is that the self-stigma of depressed individuals who had never visited psychiatric units was not measured. Finally, the translation of the SSAS from English might have limited the use of colloquial expressions for the Taiwanese population.

In conclusion, all clinicians should take the issue of self-stigma into consideration when communicating with patients with depression, especially those who have characteristics associated with high levels of self-stigma. We plan to follow the patients who participated in this study and examine the predictive value of self-stigma for medication adherence and outcomes over two years.

The authors are affiliated with the department of psychiatry of Kaoshing Medical University in Taiwan. Send correspondence to Dr. Chen at Department of Psychiatry, Kaoshing Medical University, 100 Tzyou First Road, Kaoshing City, Taiwan 807 (e-mail, [email protected]).

Enlarge table