Public Stigma in China Associated With Schizophrenia, Depression, Attenuated Psychosis Syndrome, and Psychosis-Like Experiences

This cross-sectional study of public stigma used a vignette methodology. A total of 204 Chinese participants were recruited through a public awareness campaign and through the e-mail network of the University of Hong Kong. Individuals age 18 and older who had never had a diagnosis of or treatment for a mental disorder and who consented to participate were included. The study was approved by the Institutional Review Board of the University of Hong Kong/Hospital Authority Hong Kong West Cluster and performed in accordance with ethical standards of the 1964 Declaration of Helsinki and its later amendments. Basic demographic information, including age, gender, current occupation, and education level, was collected. Previous contact with persons with mental illness was also noted on the basis of questions adapted from Alexander and Link (18). Participants read four vignettes describing individuals with APS, schizophrenia, depression, or psychosis-like experiences and responded online. They used a seven-domain stigma scale (SDSS) to rate their level of public stigma associated with each of the four individuals in the vignettes. [The four vignettes are included in an online supplement to this article.]

Each of the vignettes objectively described symptoms and functioning of the person with the disorder. The schizophrenia and depression vignettes stated the diagnosis and were previously used in study by Pescosolido and colleagues (19). The APS vignette described the likelihood of transition to psychosis as follows: “People at high risk of psychosis have not yet developed a full psychotic disorder but have some symptoms that might lead to a future psychotic disorder. However, only 35% of these individuals will go on to exhibit psychosis within 2.5 years of identification” (20). The vignette on psychosis-like experiences described a person with no psychiatric diagnosis and normal functioning who experienced transient psychotic symptoms. English versions of the Chinese vignettes were back-translated by a professional translator and proved to be satisfactory.

The SDSS is a 27-item questionnaire about public stigma adopted from Pescosolido and colleagues (19). The scale measures seven major domains of public stigma, including social distance, traditional prejudice, exclusionary sentiments, negative affect, perceptions of dangerousness, treatment carryover, and disclosure carryover. Social distance refers to reluctance to interact with members of a group. Traditional prejudice refers to adherence to the belief that all members of a group are categorically inferior to others. Exclusionary sentiment refers to the willingness to exclude people from a group from the full benefits of citizenship. Negative affect refers to public views that people from a group are difficult to interact with. Perception of dangerousness is about the fear that persons from a group represent a threat of violence to self or others. Treatment carryover is the belief that being known to have received mental health care carries long-lasting consequences. Disclosure spillover refers to the negative consequences of revealing one’s mental illness. The SDSS uses a 4-point Likert scale (1, strongly agree; 2, agree; 3, disagree; and 4, strongly disagree) to measure agreement with statements about each vignette. Each domain score was calculated by summing responses from the relevant items and dividing by the total number of items in the domain. Scores above the midpoint (2.0) indicated a stigmatizing attitude, with higher scores indicating more severe public stigma.

Statistical analyses were conducted with SPSS, version 20.0. The Kolmogorov-Smirnov test was used to test for normality of noncategorical variables, and Wilcoxon signed-rank tests were used to examine differences in scores between pairs of vignettes. Kruskal-Wallis or Mann-Whitney tests examined differences in scores between demographic variables.

As shown in Table 1, 204 participants (154 from the general public and 50 health care professionals) completed the questionnaires. The mean age of the 204 participants was 27.3. Seventy percent were female, and 64% had undergraduate or postgraduate degrees. Most participants did not have relatives or close friends who had been hospitalized in a psychiatric unit. However, about half (43%) had worked or volunteered in mental health services.

As shown in Table 2, total stigma scores were highest for schizophrenia, followed by APS, depression, and psychosis-like experiences. For both schizophrenia and psychosis-like experiences, total stigma scores and all domain scores differed significantly (p<.001) from scores for the other vignettes. For APS and depression, only the domain score for treatment carryover differed significantly.

For APS and depression, the order of domains as ranked by score was similar; the domain with the highest score for both vignettes was disclosure spillover, followed by perception of dangerousness, and treatment carryover. For APS, the two domains with the next-highest scores were social distance and traditional prejudice, followed by negative affect and exclusionary sentiments. In contrast, for depression, the two domains with the next-highest scores were traditional prejudice and social distance, also followed by negative affect and exclusionary sentiments. However, there was no significant difference between domain rankings for depression. For schizophrenia, domain rankings were as follows (from highest to lowest score): disclosure spillover, perception of dangerousness, treatment carryover, social distance, negative affect, traditional prejudice, and exclusionary sentiments. Significant differences with other vignettes were noted for social distance (Z=−2.00, p<.05) and negative affect (Z=−1.97, p<.05). For psychosis-like experiences, domain rankings were as follows: disclosure spillover, treatment carryover, traditional prejudice, social distance, dangerousness, exclusionary sentiments. Significant differences with other vignettes were noted for treatment carryover (Z=−2.09, p<.05), and exclusionary sentiments (Z=−2.11, p<.05).

Table 3 presents total public stigma scores for each vignette by demographic and other characteristics. For the APS vignette, total scores of participants from the general public were higher than scores of health care professionals (Z=−3.0, p<.01); the scores for these two groups did not differ significantly for any other vignette. For the APS vignette, some domain scores (data not shown) were also significantly higher for the public group than for the professional group: social distance (Z=−2.86, p<.01), negative affect (Z=−3.98, p<.001), and perceptions of dangerousness (Z=−2.39, p<.05). No significant differences between rankings were found by gender and age for any vignette.

Significant differences in total stigma scores were found by education level for both the APS and the schizophrenia vignettes. For the APS vignette, participants who had a high school diploma had significantly higher scores than participants in the other education-level groups (χ²=12.8, df=4, p<.05). For the APS vignette, scores in two domains (data not shown) were also significantly higher for those with a high school diploma than for those in the other education-level groups: traditional prejudice (χ²=11.7, df=4, p<.05) and negative affect (χ²=13.0, df=4, p<.05). For the schizophrenia vignette, participants without a high school diploma (secondary level) had significantly higher scores than those in the other education-level groups on one domain: treatment carryover (χ²=11.1, df=4, p<.05).

For all vignettes, total stigma scores of participants who reported working or volunteering in mental health services were significantly lower than total scores of participants who reported no such experiences (Table 3). For the APS vignette, not only did the total stigma score differ between these groups (χ²=24.1, df=2, p<.001), but the scores on five domains (data not shown) also differed: social distance (χ²=25.7, df=2, p<.001), traditional prejudice (χ²=10.2, df=2, p<.01), negative affect (χ²=26.0, df=2, p<.001), perception of dangerousness (χ²=13.2, df=2, p<.001), and treatment carryover (χ²=8.6, df=2, p<.01). For the schizophrenia vignette, scores between the general public and health care professionals differed for two domains: negative affect (χ²=8.0, df=2, p<.05) and treatment carryover (χ²=9.8, df=2, p<.01). For the depression vignette, scores between the two groups differed only for the domain of treatment carryover (χ²=8.9, df=2, p<.05). For the psychosis-like experiences vignette, scores between the two groups differed for two domains: exclusionary sentiments (χ²=8.5, df=2, p<.05) and negative affect (χ²=8.6, df=2, p<.01).

Finally, for the APS vignette, participants who reported encountering someone in public who appeared to be mentally ill either often or sometimes had higher scores on two domains than those who reported almost never having such encounters (data not shown): exclusionary sentiments (χ²=1.0, df=3, p<.05) and negative affect (χ²=15.3, df=3, p<.01).