The prevalence and predictors of self-stigma of individuals with mental health illness in two Chinese cities

Abstract

Background:

Although self-stigma is found to have adverse effects on the lives of persons with mental illness, little is known on the self-stigma of these individuals in Chinese societies.

Objective:

This research study explores the prevalence rate and predicting factors of self-stigma of consumers in two Chinese cities, Hong Kong and Guangzhou.

Methods:

A cross-sectional research design is adopted which involves a random sample of 266 consumers from Hong Kong and a convenient sample of 208 consumers from Guangzhou. These individuals have been assessed in terms of their self-stigma, recovery, self-esteem and quality of life by using standardized assessment scales.

Results:

In all, 38.3% of the Hong Kong participants and 49.5% of the Guangzhou participants report to have self-stigma. Also, self-stigma is found to be negatively related to self-esteem and quality of life. A logistic regression analysis shows that hope and well-being are predicting factors of self-stigma.

Conclusion:

Self-stigma is found to be higher in Guangzhou, probably due to the influence of traditional cultural values. Also, as hope and well-being are found to be predicting factors of self-stigma, suitable recovery-orientated interventions that facilitate hope and well-being should be developed so as to reduce self-stigma of consumers in Chinese societies.

Keywords

Self-stigma recovery mental illness Chinese culture

Introduction

In Chinese societies, it is fairly common for persons with mental illness to experience public stigma. For example, local surveys in Hong Kong reported that more than half of mental health consumers felt that they had experienced public stigma and discrimination (Chung & Wong, 2004; Lee, Lee, Chiu, & Kleinman, 2005). Also, stigmatizing attitudes are found to be common among Hong Kong community residents (Tsang, Tam, Chan, & Cheung, 2003), employers (Tsang et al., 2007), family caregivers (Lee et al., 2005), and mental health professionals (Chien, Yeung, & Chan, 2014; Lee, Chiu, Tsang, Chui, & Kleinman, 2006), such that persons with mental illness are perceived as quick-tempered, unpredictable, dangerous, abnormal, weird, introverted, and having low self-esteem (Chien et al., 2014; Lee et al., 2005; Tsang et al., 2007; Tsang et al., 2003). One consequence of public stigma is that consumers often internalize the stigmatizing beliefs, and ultimately self-stigmatize themselves (Corrigan, Watson, & Barr, 2006).

The problem of self-stigmatization is common among consumers in both Chinese and Western societies. Much research studies have been done on self-stigma in the Western societies. Recent research studies reported that 41.7% of consumers in Europe (Brohan, Elgie, Sartorius, & Thornicroft, 2010) and 36.1% of consumers in the United States (West, Yanos, Smith, Roe, & Lysaker, 2011) had self-stigma. Self-stigma is found to have negative impacts on the consumer’s life, such as reducing self-esteem, self-efficacy, and quality of life, while increasing psychiatric symptoms in the Western societies (Boyd, Adler, Otilingam, & Peters, 2014; Corrigan et al., 2006; Gerlinger et al., 2013; Livingston & Boyd, 2010; Ritsher, Otilingam, & Grajiales, 2003) and the Chinese societies (Fung, Tsang, Corrigan, Lam, & Cheng, 2007; Tang & Wu, 2012; Young, Ng, Pan, & Cheng, 2015). Also, research shows that self-stigma is not predicted by socio-demographic factors such as gender, education, employment and diagnosis, but predicted by psychosocial variables such as hope and empowerment (Gerlinger et al., 2013; Livingston & Boyd, 2010).

The research studies on self-stigma outlined above have some limitations. First, research study on the prevalence rate of self-stigma in the non-Western context, such as the Chinese societies, is lacking. Self-stigmatization is not only influenced by the severity of the symptoms and functioning level of an individual but also by the socio-cultural environment (Fabrega, 1991; Lam et al., 2010; Lee et al., 2005; Yang, 2007). Thus, it is important to study the prevalence rate of self-stigma and identify factors that would predict self-stigma in the non-Western context, such as the Chinese societies, so as to facilitate cross-cultural studies on self-stigma of persons with mental illness. Second, the identification of factors predicting self-stigma remains inconclusive. For example, the research findings on predicting role of demographic factors such as gender, education, employment and diagnosis are mixed in different cultures including Chinese culture (Drapalski et al., 2013; Lv, Wolf, & Wang, 2013). Third, very few research studies have explored the predicting role of recovery on self-stigma (Boyd et al., 2014). Researchers have suggested that self-stigma is negatively related to and predicted by hope and empowerment (Gerlinger et al., 2013; Livingston et al., 2010), which are important elements of recovery (Anthony, 1993; Davidson, Sells, Sangster, & O’Connell, 2005; Leamy, Bird, Le Boutillier, Williams, & Slade, 2011). However, recovery can also consist of other elements in addition to hope and empowerment. For example, disability management, well-being, autonomy, stop medication and stable open employment have been identified as other important recovery elements for consumers in Chinese societies (Chiu, Ho, Lo, & Yiu, 2010; Ng et al., 2008; Song & Hsu, 2011). Yet, whether recovery is really able to predict self-stigma, and if so, which recovery elements predict self-stigma for consumers, remains unanswered. Thus, more research work is needed on identifying predicting factors of self-stigma, especially in non-Western societies such as Chinese societies.

In Hong Kong and other Chinese societies, study on the self-stigmatization of persons with mental illness is lacking. Therefore, this research study attempts to (a) explore and compare the prevalence of self-stigmatization among consumers who were participating in community-based mental health services provided by local non-governmental organizations (NGOs) in two Chinese cities, that is, Hong Kong and Guangzhou and (b) identify factors that predict self-stigma of these persons in these two Chinese cities.

Although both Hong Kong and Guangzhou cities have been deeply influenced by traditional Chinese cultural values, both cities have very different socio-cultural conditions (Lam, Ng, Pan, & Young, 2015; Tse, Ran, Huang, & Zhu, 2013; Xiang, Yu, Sartorius, Ungvari, & Chiu, 2008). Hong Kong is a highly modernized and Westernized city with more comprehensive community-based mental health services, while Guangzhou is more traditional and its development of community-based mental health services is still in an early stage with limited resources such that persons with mental illness rely heavily on the family care and support (Lam et al., 2015; Xiang et al., 2008). In this study, it is hypothesized that mental health consumers in Hong Kong and Guangzhou (a) have a different prevalence rate and (b) have different predicting factors of self-stigma. The research results will facilitate comparison studies on the self-stigmatization of consumers across various cultures and societies.

Methods

Procedure

This study adopts a cross-sectional research design which involves a random sample of 266 consumers in Hong Kong and a convenient sample of 208 consumers in Guangzhou during the period of 2013–2014. The sampling methods in Hong Kong and Guangzhou were different due to various practical reasons, including limitations in accessing consumers in the two cities, different levels of support and involvement of the local NGOs with this study between the two cities and so on. The Hong Kong sample was randomly recruited from a population of consumers who are receiving community-based mental health services provided by four local NGOs, which offer a variety of services and provide different facilities, including five psychiatric halfway houses, two sheltered workshops with supported employment services and five community mental health centers. These serve over 6000 consumers. By using the stratified sampling method (Särndal, Swensson, & Wretman, 2003), a random sample was recruited from each of these service units in Hong Kong. The Guangzhou sample was recruited with a convenient sampling method (Huck, 2000) from a population of consumers who are receiving services from two local NGOs which provide comparatively limited mental health services, including a sheltered workshop, a family resource center and two integrated family service centers.

Inclusion criteria

The inclusion criteria for this study include (a) meeting the criteria cited in the Diagnostic and Statistical Manual of Mental Disorders–Fourth Edition (DSM-IV-TR) (American Psychiatric Association, 2005) for any diagnosis of a mental disorder, (b) aged 18 years or above and (c) receiving one or more community-based mental health services from local NGOs. Those who have a primary or comorbid diagnosis of intellectual or learning disability were excluded.

A questionnaire survey was conducted with the participants. Demographic and clinical characteristics were collected from the participants and/or from the service units involved. The majority of the questionnaires were self-administered, whereas questions were read out to the few illiterate participants by the researcher.

Measuring scales

The Internalized Stigma of Mental Illness (Chinese version) (ISMI-C) (Young et al., 2015), originally developed by Ritsher et al. (2003), is a 24-item self-administered questionnaire designed to measure the subjective experience of self-stigma, and consists of subscales that measure shame/alienation, stereotype endorsement, perceived discrimination and social withdrawal. ISMI has been widely used internationally with satisfactory reliability and validity across various cultures (Boyd et al., 2014). The construct validity of the Chinese version of ISMI-C (24-items) has been confirmed, with good internal consistency (α = .93) (Young et al., 2015). Each item is rated on a 4-point scale that ranges from 1 = strongly disagree to 4 = strongly agree. As suggested by previous studies (Brohan et al., 2010; Ritsher et al., 2003), the cutoff point of the ISMI is 2.5, which is also adopted in the present study. In the present sample, the internal consistency of the ISMI-C (24 item) (α = .94) was high.

The Stages of Recovery Scale (Chinese scale) (SOR) (Song & Hsu, 2011) is used in this study for assessing recovery in the Chinese societies. Also, in the present study, Anthony’s (1993) approach on recovery, which focuses on the recovery process, is used, and recovery is defined as a way of living a satisfying, hopeful and contributing life even with the limitations caused by mental illness. The SOR consists of 45 items that assess six elements of recovery, including regaining autonomy, disability management, overall well-being, better social functioning, sense of hope and helping others. Each item is rated on a 4-point scale that ranges from 1 = strongly disagree to 4 = strongly agree. The construct validity of the Chinese version of the SOR has been confirmed, with good internal consistency (α = .80–.95) and test–retest reliability (r = .72) (Song & Hsu, 2011). In the present study, the internal consistency of the entire SOR scale was high (α = .96).

The Rosenberg Self-Esteem Scale (RSES) has been widely used in many countries and translated into many languages, including Chinese (Schmitt & Allik, 2005). The RSES consists of 10 items. Each item is rated on a 4-point scale that ranges from 1 = strongly disagree to 4 = strongly agree. The reliability and validity of the Chinese version of the RSES have been shown to be acceptable (α = .63) (S. O. Leung & Wong, 2008), and it was used in the present study. In the present sample, the internal consistency of the RSES was satisfactory (α = .74).

The World Health Organization (WHO) Quality of Life Instrument (WHOQOL-BREF) (Chinese version) has been reported to demonstrate satisfactory estimates of reliability and validity, and allows for cross-cultural sensitivity (Skevington, Lotfy, & O’Connell, 2004). The Hong Kong Chinese version of the WHOQOL-BREF has been found to have satisfactory validity, internal consistency ranging from Cronbach’s α = .59 to .78, test–retest reliability and inter-rater reliability in all domains with Cronbach’s α = .75 and above (K. F. Leung, Wong, Tay, Chu, & Ng, 2005). In this study, two items (‘How would you rate your quality of life?’ and ‘How satisfied are you with your health?’) are adopted from the Chinese version of the WHOQOL-BREF (HK) to measure the overall quality of the life of the respondents. Each item is rated on a 5-point scale that ranges from 1 = strongly disagree to 5 = strongly agree.

Statistical analysis

Statistical analysis was performed by using SPSS 22.0 (IBM Corp., 2013). Any values of p < .05 were considered statistically significant. The baseline demographic and clinical characteristics between these two cities (i.e. Hong Kong and Guangzhou) were compared by using one-way analysis of variance (ANOVA) for the continuous variables and chi-squared (χ²) tests for the categorical variables.

For each city, the scores on the ISMI and its subscales were calculated and presented as the total scores, mean scores and standard deviations (SDs). The normality of the data was checked. With the use of the one-way analysis of covariance (ANCOVA) which is a general linear model, the total scores on the ISMI and its subscales between the two cities were compared.

To identify the variables that are related to self-stigma, the demographic and psychosocial variables of the self-stigmatized group (ISMI mean score ⩾ 2.5) were compared with those of the non-self-stigmatized group in the two cities by using one-way ANOVA for the continuous variables and chi-squared (χ²) tests for the categorical variables.

In order to identify the predicting variables for the ISMI, univariate and multivariate logistic regression analyses were conducted, with the ISMI included as a dependent variable, and their related demographic and psychosocial variables were placed into multivariate logistic regression models. Stepwise model reduction was conducted by dropping any non-significant variables from the regression model.

Ethical considerations

The ethical issues related to this study were taken into consideration, and the study was approved by the Review Board of the Executive Committee of the NGOs involved. All of the participants provided written informed consent.

Results

A total of 474 participants (Hong Kong n = 266; Guangzhou n = 208) were included in the analyses in which 219 were male (46.2%) and 255 were female (53.8%). The average age of the participants was 42.61 years (SD = 10.70 years). Participants from the two cities were significantly different in their demographic and clinical characteristics (see Table 1). Most participants of the Hong Kong sample were female (62.4%), single (62.0%), schizophrenic (62.1%), had attained an education to the secondary school level (71.4%), unemployed or housewives (45.5%) and dependent on government income support for livelihood (93.1%). In comparison to the Hong Kong sample, the Guangzhou sample had a higher percentage of participants who are male (57.2%), younger in age (mean age of 40.9; SD = 9.70), received a diagnosis of schizophrenia (81.0%), working at sheltered workshops (50.5%), living with family members (91.7%) and relying on family financial support (53.2%).

Table 1.

Demographic and clinical characteristics of the participants in the present study.

Characteristics	Valid data	Hong Kong (N = 266), n (%)	Guangzhou (N = 208), n (%)	Total (N = 474), n (%)	p value
Sex	474				.000**^a
Male		100 (37.6)	119 (57.2)	219 (46.2)
Female		166 (62.4)	89 (42.8)	255 (53.8)
Diagnosis	440				.000**^a
Schizophrenia		149 (62.1)	162 (81.0)	311 (70.7)
Depression/anxiety		86 (35.8)	28 (14.0)	114 (25.9)
Others		5 (2.1)	10 (5.0)	15 (3.4)
Marital status	474				.001**^a
Single		165 (62.0)	144 (69.2)	309 (65.2)
Married		41 (15.4)	44 (21.2)	85 (17.9)
Divorce, separated, widowed		48 (18.1)	19 (9.1)	67 (14.1)
Others		12 (4.5)	1 (0.5)	13 (2.7)
Education level	474				.000**^a
University		33 (12.4)	45 (21.6)	78 (16.4)
Secondary		190 (71.4)	124 (59.6)	314 (66.2)
Primary		40 (15.1)	20 (9.6)	60 (12.7)
None		3 (1.1)	19 (9.2)	22 (2.7)
Place of living	464				.000**^a
Own home		147 (56.7)	200 (97.5)	352 (74.8)
Half-way house		95 (36.7)	0 (0.0)	95 (20.5)
Residential home		17 (6.6)	5 (2.5)	22 (4.7)
Living arrangement	466				.000**^a
With family		116 (43.3)	188 (91.7)	301 (64.6)
Alone		36 (13.8)	15 (7.3)	51 (10.9)
Others (residential home)		112 (42.9)	2 (1.0)	144 (24.5)
Occupational status	470				.000**^a
OE (FT/PT)		52 (19.7)	14 (6.8)	66 (14.0)
SW/SE		92 (34.8)	104 (50.5)	196 (41.7)
Unemployed, housewife, retired		120 (45.5)	88 (42.7)	208 (44.3)
Main source of income	462				.000**^a
Employment		43 (16.5)	28 (13.9)	71 (15.4)
Support from family		37 (14.1)	107 (53.2)	144 (31.2)
Government fin. support		180 (69.0)	49 (24.4)	229 (49.6)
Others		1 (0.4)	17 (8.5)	18 (3.9)
		M (SD)	M (SD)
Age	469	43.9 (11.3)	40.9 (9.7)	42.61 (10.7)	.003**^b
Period of illness	437	14.5 (10.1)	17.6 (9.3)	15.9 (9.8)	.001**^b
Number of hospitalization	446	2.8 (3.1)	1.9 (1.9)	2.4 (2.6)	.000**^b

OE (FT/PT): open employment (full time/part time); fin. support: financial support. SW/SE: sheltered workshop/supported employment; SD: standard deviation.

Chi square.

Analysis of variance (ANOVA).

Significant at p < . 05; **significant at p < . 01.

Prevalence rate of self-stigma between Hong Kong and Guangzhou respondents

Table 2 gives the prevalence rate of self-stigma of the consumers in Hong Kong and Guangzhou, with 38.3% of the Hong Kong participants and 49.5% of the Guangzhou participants who reported self-stigma (i.e. ISMI mean score ⩾ 2.5).

Table 2.

Comparison of ISMI scores between Hong Kong (HK) and Guangzhou (GZ).

	n (%)	HK (N = 266)	n (%)	GZ (N = 208)	n (%)	Total (N = 474)	F	ANCOVA
		Mean (SD)		Mean (SD)		Mean (SD)		p
ISMI (24 item)	102 (38.3)	2.34 (0.48)	103 (49.5)	2.50 (0.54)	205 (43.2)	2.42 (0.52)	15.11	.000**
Shame	139 (52.3)	2.44 (0.59)	111 (53.4)	2.57 (0.63)	250 (52.7)	2.51 (0.62)	4.99	.026*
Stereotype endorsement	62 (23.3)	2.21 (0.47)	84 (40.4)	2.43 (0.60)	146 (30.8)	2.30 (0.55)	9.01	.003**
Perceived discrimination	110 (41.4)	2.37 (0.61)	113 (54.3)	2.53 (0.60)	223 (47.0)	2.45 (0.62)	11.87	.001**
Social withdrawal	123 (46.2)	2.37 (0.61)	109 (52.4)	2.51 (0.65)	232 (48.9)	2.44 (0.63)	8.84	.003**

SD: standard deviation; ANCOVA: analysis of covariance; ISMI: Internalized Stigma of Mental Illness.

Significant at p < .05; **significant at p < . 01.

Also, by using the ANCOVA analysis, the scores of ISMI and its four subscales between the Hong Kong and Guangzhou samples were compared after controlling for the effects of all the demographic variables, including sex, age, educational level, marital and occupational statuses, main source of income, living arrangements, diagnosis, period of illness and number of hospitalizations (see Table 2). The results show that the scores of ISMI as well as its four subscales of the Hong Kong sample were significantly lower than those of the Guangzhou sample.

On the other hand, among the four subscales of the ISMI, both the Hong Kong and Guangzhou respondents provided shame/alienation with the highest ranking (M = 2.51, SD = .62), followed by perceived discrimination (M = 2.45, SD = 0.62) and social withdrawal (M = 2.44, SD = 0.63), while stereotype endorsement was ranked the lowest (M = 2.30, SD = 0.62).

Factors related to self-stigma

In all but one of the demographic variables (sex), the self-stigmatized group does not significantly differ from the non-self-stigmatized group in the Hong Kong, Guangzhou and total samples. In the Hong Kong sample, more females (70.6%) than males (29.4%) have self-stigma (χ² = 4.721, p = .037<.05). However, this difference is not found in the Guangzhou sample (see Table 3). These findings suggest that almost all of the demographic variables are not related to the ISMI.

Table 3.

Correlation among self-stigma, self-esteem, quality of life and recovery in the Hong Kong, Guangzhou and total samples.

	Self-stigmatized group, Hong Kong sample (N = 102)		Self-stigmatized group, Guangzhou sample (N = 103)		Self-stigmatized group, total sample (N = 205)
	χ²	p value^a	χ²	p value^b	χ²	p value^c
Sex^d	4.721	.037*	0.000	1.000 (n.s.)	1.130	.307 (n.s.)
Diagnosis^d	0.017	.992 (n.s.)	2.374	.305 (n.s.)	1.990	.370 (n.s.)
Marital status^d	2.727	.436 (n.s.)	1.152	.764 (n.s.)	2.013	.570 (n.s.)
Education level^d	1.344	.719 (n.s.)	5.816	.213 (n.s.)	2.966	.560 (n.s.)
Place of living^d	5.148	.161 (n.s.)	.358	.620 (n.s.)	0.540	.463 (n.s.)
Living arrangement^d	2.622	.270 (n.s.)	2.039	.361 (n.s.)	0.186	.911 (n.s.)
Occupational status^d	3.789	.285 (n.s.)	5.704	.127 (n.s.)	7.212	.065 (n.s.)
Main source of income^d	6.911	.075 (n.s.)	4.314	.299 (n.s.)	7.590	.055 (n.s.)
	F	p value^a	F	p value^b	F	p value^c
Age^e	0.396	.530 (n.s.)	0.197	.657 (n.s.)	0.194	.660 (n.s.)
Period of illness^e	0.185	.667 (n.s.)	1.408	.237 (n.s.)	2.154	.143 (n.s.)
Number of hospitalization^e	3.015	.084 (n.s.)	0.421	.517 (n.s.)	1.865	.173 (n.s.)
RSES^e	63.553	.000**	74.227	.000**	123.726	.000**
Overall QoL^e	27.985	.000**	6.183	.014*	28.263	.000**
SOR total score^e	10.083	.002**	1.769	.003**	15.138	.000**
Hope^e	11.100	.001**	1.623	.009**	20.197	.000**
Dis. management^e	1.208	.273 (n.s.)	1.413	.046*	2.811	.094 (n.s.)
Autonomy^e	9.151	.003**	1.654	.007**	14.130	.000**
Functioning^e	3.432	.065 (n.s.)	1.697	.005**	4.729	.030*
Well-being^e	17.902	.000**	1.693	.005**	22.442	.000**
Helping others^e	5.822	.017*	1.139	.259 (n.s.)	8.740	.003**

SOR: Stages of Recovery Scale; RSES: Rosenberg Self-Esteem Scale; QoL: WHOQOL-BREF – World Health Organization Quality of Life Instrument; Dis. management: Disability management.

Compare with non-self-stigmatized group in Hong Kong sample.

Compare with non-self-stigmatized group in Guangzhou sample.

Compare with non-self-stigmatized group in total sample.

Pearson chi-square.

Analysis of variance (ANOVA).

Significant at p < .05; **significant at p < . 01; n.s.: non-significant.

On the other hand, results of ANOVA indicate that the self-stigmatized group has significantly lower scores of RSES, WHOQOL-BREF, SOR and its five subscales, including hope, autonomy, functioning, well-being and helping others than that of non-self-stigmatized group in the Hong Kong, Guangzhou and total samples (see Table 3). Pearson correlational analyses further show that the ISMI score is negatively and moderately related to the RSES score in the Hong Kong (r = −.47, p < .001), Guangzhou (r = −.58, p < .001) and total (r = −.49, p < .001) samples. Also, the ISMI score is found to be negatively and mildly to moderately related to the WHOQOL-BREF score in the Hong Kong (r = −.35, p < .001), Guangzhou (r = −.18, p < .001) and total (r = −.27, p < .001) samples. Furthermore, the ISMI and its five subscales scores are negatively and mildly to moderately related to the SOR scores as well as its five subscales in the Hong Kong (r ranked from −.14 to −.32, p < .05), Guangzhou (r ranked from −.14 to −.32, p < .05) and total (r ranked from −.09 to −.23, p < .05) samples.

Factors that predict self-stigma

As the Hong Kong, Guangzhou and the total samples share very similar variables that predict ISMI, the results of the univariate and multivariate logistic regression analyses are presented only for the total sample (see Tables 4 and 5).

Table 4.

Univariate logistic regression analysis on ISMI and its subscales in the total sample (N = 474).

	Self-stigma		Shame		Stereotype endorsement		Perceived discrimination		Social withdrawal
	OR	95% CI	OR	95% CI	OR	95% CI	OR	95% CI	OR	95% CI
Sex	.820	0.569–1.182	1.481*	1.030–2.130	0.869	0.588–1.284	1.187	0.826–1.705	1.114	0.776–1.600
Age	1.004	0.987–1.021	1.007	0.990–1.024	1.009	0.991–1.028	1.006	0.989–1.024	.995	0.978–1.012
Diagnosis of Schi	1.234	0.840–1.814	.962	0.658–1.407	1.205	0.795–1.827	1.336	0.912–0.958	1.290	0.882–1.887
No. of hospitalizations	1.049	0.979–1.125	1.031	0.961–1.105	1.010	0.938–1.087	1.036	0.966–1.110	1.036	0.967–1.111
Open employment	.664	0.386–1.143	.712	0.422–1.200	0.750	0.416–1.353	.696	0.410–1.183	0.850	0.504–1.432
Income from F/G	1.719*	1.007–2.933	1.066	0.687–1.655	0.486	0.529–1.351	1.136	0.730–1.767	1.189	0.765–1.849
SOR total score	.985**	0.978–0.992	.984**	0.976–0.991	0.987**	0.979–0.996	.988**	0.980–0.995	0.983**	0.975–0.990
Hope	.820**	0.750–0.897	.864**	0.791–0.944	0.867*	0.791–0.950	.888**	0.814–0.968	0.782**	0.713–0.858
Dis. management	.972	0.939–1.005	.974	0.941–1.007	0.964*	0.930–1.000	.979	0.947–1.013	0.960*	0.928–0.993
Autonomy	.962**	0.942–0.982	.985**	0.938–0.978	0.970**	0.949–0.991	.968**	0.949–0.988	0.962**	0.943–0.982
Functioning	.953*	0.912–0.996	.939**	0.899–0.982	0.953**	0.909–0.999	.948*	0.907–0.991	0.930**	0.889–0.972
Well-being	.919**	0.887–0.953	.896**	0.863–0.930	0.955**	0.920–0.991	.935**	0.902–0.968	0.908**	0.875–0.942
Helping others	.880**	0.807–0.959	.922	0.846–1.004	0.827**	0.755–0.907	.904*	0.830–0.985	0.888**	0.815–0.968

ISMI: Internalized Stigma of Mental Illness; Schi: schizophrenia; Income from F/G: income manly rely on family or government financial support; SOR: Stage of Recovery Scale; Dis. management: disability management; OR: odds ratio; CI: confidence interval.

Significant at p < .05; **significant at p < . 01.

Table 5.

Multivariate logistic regression analysis on ISMI and its subscales in the total sample (N = 474).

Multivariate logistic regression model	OR	95% CI		p value
Dependent variable: Self-stigma
Hope	.879	.795	.973	.013*
Well-being	.942	.904	.981	.004**
Model χ²	28.029 (p < .001)
Nagelkerke R²	.077
Dependent variable: Shame
Well-being	.896	.863	.930	.000**
Model χ²	35.680 (p < .001)
Nagelkerke R²	.097
Dependent variable: Stereotype endorsement
Helping Others	.827	.755	.907	.000**
Model χ²	16.741 (p < .001)
Nagelkerke R²	.049
Dependent variable: Perceived discrimination
Well-being	.935	.902	.968	.000**
Model χ²	14.454 (p < .001)
Nagelkerke R²	.040
Dependent variable: Social withdrawal
Hope	0.763	.672	.867	.000**
Disability Management	1.074	1.021	1.130	.006**
Well-being	0.918	.878	.960	.000**
Model χ²	46.822 (p < .001)
Nagelkerke R²	.125

ISMI: Internalized Stigma of Mental Illness; Dis. management: disability management; OR: odds ratio; CI: confidence interval.

Significant at p < .05; **significant at p < . 01.

The multivariate logistic regression analysis indicates that internalized stigma is significantly predicted by hope (odds ratio (OR) = .879, 95% confidence interval (CI): .801–.983, p < .05) and well-being (OR = .942, 95% CI: .903–.980, p < .001). Shame is significantly predicted by well-being (OR = .896, 95% CI: .863–.930, p < .001). Stereotype endorsement is significantly predicted by helping others (OR = .827, 95% CI: .755–.907, p < .001). Perceived discrimination is significantly predicted by well-being (OR = .935, 95% CI: .902–.968, p < .001). Social withdrawal is significantly predicted by hope (OR = .763, 95% CI: .672–.867, p < .001), disability management (OR = 1.074, 95% CI: 1.021–1.130, p < .01) and well-being (OR = .918, 95% CI: .878–.960, p < .001).

Discussion

In this study, 38.3% of the Hong Kong participants and 49.5% of the Guangzhou participants are found to have self-stigma. Taken together, 43.2% of consumers in Hong Kong and Guangzhou have self-stigma. These prevalence rates of self-stigma are comparable to the 41.7% reported in Europe (Brohan et al., 2010) and 36.1% reported in the United States (West et al., 2011). This finding suggests that self-stigma is common among consumers in Hong Kong, Guangzhou and Western cities.

Consistent with the hypothesis (1), in the present study, Guangzhou consumers were found to have a significant higher rate of self-stigma than their Hong Kong counterparts, even after controlling for their differences with all demographic characteristics, which is probably due to the different socio-cultural conditions involved (Brohan et al., 2010; Fabrega, 1991; Lam et al., 2010; Lee et al., 2005; Yang, 2007) and warrants more discussion. As mentioned earlier, the development of community mental health service in Guangzhou is still in an early stage with limited resources such that persons with mental illness rely heavily on family care (Lam et al., 2015; Tse et al., 2013). Thus, as shown in the present study, as compared to Hong Kong consumers, more (91.7%) Guangzhou consumers are living with their families, more (53.2%) rely on financial support from family members, but fewer (6.8%) achieve open employment. Although traditional Chinese values emphasize collectivism and that the family is responsible for the care of consumers with mental illness, many of their family caregivers experience public stigma for having a relative who is suffering from mental illness, and subsequently experience ‘shame’ and ‘loss of face’ within the community (Lam et al., 2010; Lee et al., 2005; Yang, 2007). The public stigma experienced by the Chinese family caregivers increases their caring burden, distress and level of expressed emotions (Mak & Cheung, 2012; Philips, Pearson, Li, Xu, & Yang, 2002), which in turn leads to their rejection and stigmatizing attitudes toward their relatives who are suffering from mental illness (Chien et al., 2014; Lee et al., 2005). Moreover, as persons with mental illness rely on the family for financial support to sustain their everyday life, they are regarded by their families and communities as ‘dependent’ and constituting as a ‘burden’ on their families (Hsiao, Klimidis, Minas, & Tan, 2005; Lam et al., 2010; Yang, 2007). Under the influence of such traditional cultural values and negative attitude of family members, a higher rate of self-stigma is found among Guangzhou consumers than their Hong Kong counterparts. It seems that the more they rely on family care, the more public and self-stigmatization Chinese consumers experience (Lee et al., 2005).

On the other hand, in contrast to hypothesis (2), in this study, Hong Kong and Guangzhou consumers are found to share very similar experiences in self-stigma and predicting factors of self-stigma, suggesting that although Hong Kong and Guangzhou are different in their socio-cultural conditions, consumers of both cities are deeply influenced by traditional Chinese culture on their self-stigmatization. In particular, this study finds that shame/alienation is ranked the highest among the four components of self-stigma, indicating that shame is the most important and common element of self-stigma in the two Chinese cities, which can be explained by traditional Chinese cultural values. Due to the stigmatization toward mental illness by traditional Chinese cultural values, especially Confucianism, persons with mental illness are perceived as those who are not competent members of society, fail to comply with the five cardinal relations set out by Confucianism and subsequently are those who cause ‘loss of face’ and ‘shame’ within the family and social systems (Hsiao et al., 2005; Lam et al., 2010; Yang, 2007). In addition, the concept of ‘fate’, which reflects the acceptance of the unchangeable destiny of one’s role as set by the Transcendence, facilitates the internalization of public stigma and development of self-stigma (Lam et al., 2010). Under such cultural influences, persons with mental illness are likely to endorse public stigma, develop self-stigma and subsequently experience ‘shame’, ‘loss of face’ and ‘devaluation’ within their families and communities.

Also, in the present study, Hong Kong and Guangzhou consumers are found to share very similar factors that relate to and predict their self-stigma. In particular, self-stigma is found to be negatively related to the overall quality of life, self-esteem and recovery of consumers in both Hong Kong and Guangzhou cities. These findings are consistent with the previous studies done in the Chinese and Western societies (Chiu et al., 2010; Corrigan et al., 2006; Fung et al., 2007; Gerlinger et al., 2013; Livingston et al., 2010; Ritsher et al., 2003; Tang & Wu, 2012; Young et al., 2015), which suggests that self-stigma has adverse effects on the lives of consumers in Hong Kong, Guangzhou and Western cities.

Moreover, this study finds that in Hong Kong and Guangzhou cities, self-stigma as well as all four of its related components, including shame, stereotype endorsement, perceived discrimination and social withdrawal, is significantly related to and predicted by various recovery elements, but not demographic variables. In particular, self-stigma is predicted by hope and well-being. Hope of recovery and well-being/having a meaningful and satisfying life have been identified as important components of recovery in Chinese and Western cultures (Anthony, 1993; Chiu et al., 2010; Corrigan, Salzer, Ralph, Sangster, & Keck, 2004; Davidson et al., 2005; Leamy et al., 2011; Song & Hsu, 2011), but few research studies have been done to explore the predictive role of hope and well-being on self-stigma (Boyd et al., 2014). The identification of recovery elements as predicting factors for self-stigma certainly has important implications for community-based mental health services. In order to reduce the self-stigma of consumers, mental health professionals should develop and implement suitable recovery-orientated interventions that facilitate hope and well-being for consumers. In fact, this notion is supported by a pilot research study carried out in the United States, which reported the effective reduction of the self-stigma of persons with mental illness in a recovery-orientated program at a day clinic (Sibitz, Provaznikova, Lipp, Lakeman, & Amering, 2013). The program components of the recovery-orientated program consisted of an educational program instilling hope of recovery, counter self-devaluation, and pursuing of meaningful and satisfying lives even with the presence of psychiatric symptoms. Nevertheless, more research studies are needed to further validate the effectiveness of recovery-oriented interventions on reducing the self-stigma of consumers, especially in the Chinese culture and societies.

On the other hand, this study finds that the prevalence rates of Hong Kong and Guangzhou differ significantly, supporting the notion that self-stigma is affected by socio-cultural conditions (Fabrega, 1991; Lam et al., 2010; Lee et al., 2005; Yang, 2007). Thus, it is important to develop favorable socio-cultural conditions and mental health systems to reduce public and self-stigma (Brohan et al., 2010; Evans-Lacko, Brohan, Mojtabai, & Thornicroft, 2012). Promising mass interventions have been suggested to facilitate illness disclosure, positive social contacts, access to care and help seeking (Evans-Lacko et al., 2012).

Several methodological limitations of this study require attention. First, although the random sampling method is adopted in the Hong Kong sample, the convenience sampling method is used with the Guangzhou sample due to various practical limitations and difficulties. Although Guangzhou sample consists of a large sample size, it is probably better to also use the random sampling method for the Guangzhou sample, too. Second, the generalizability of the research results is limited by the studied sample representing those Chinese consumers who participate in community-based mental health services provided by local NGOs. Therefore, more research studies are needed in the future to investigate the self-stigma of Chinese consumers with different clinical and demographical characteristics. Third, the SOR is used in this study to assess six recovery elements, but it does not include other recovery elements such as those recovery elements as assessed by the Recovery Assessment Scale, including willingness to ask for help, goal and success orientation, reliance on others and no domination by symptoms (Corrigan et al., 2004; McNaught, Caputi, Oades, & Deane, 2007). It is therefore recommended that different assessment scales on recovery are used in future studies to validate the predicting role of recovery on self-stigma. Finally, a causal inference cannot be made for the cross-sectional research design adopted in this study. In future, a longitudinal study should be conducted to test and verify the research results.

Conclusion

In the present study, 38.3% of the Hong Kong consumers and 49.5% of the Guangzhou consumers reported to have self-stigma, thus indicating that self-stigma is very common among consumers in Chinese societies. Also, self-stigma is found to be significantly higher in Guangzhou, which is probably due to the different socio-cultural conditions involved in these two cities. Moreover, self-stigma is found to be negatively related to the overall quality of life and self-esteem, suggesting that self-stigma has adverse effects on the lives of consumers. Furthermore, this study shows that self-stigma as well as four of its components is predicted by various recovery elements. In particular, self-stigma is predicted by well-being and hope. Thus, suitable recovery-orientated interventions that facilitate hope and well-being should be developed in order to reduce the self-stigma of consumers in Hong Kong, Guangzhou and other Chinese societies.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was generously fully sponsored by the Hong Kong Baptist University [Cost Centre: 38-40-094]

References

American Psychiatric Association. (2005). Psychiatric research report. Washington, DC: Author.

Anthony

W. A.

(1993). Recovery from mental illness: The guiding vision of the mental health service system in the 1990s. Psychosocial Rehabilitation Journal, 16, 11–23.

Boyd

J. E.

Adler

E. P.

Otilingam

P. G.

Peters

(2014). Internalized stigma of mental illness (ISMI) scale: A multinational review. Comprehensive Psychiatry, 55, 221–231. doi:10.1016/j.comppsych.2013.06.005

Brohan

Elgie

Sartorius

Thornicroft

(2010). Self-stigma, empowerment and perceived discrimination among people with schizophrenia in 14 European countries: The GAMIAN-Europe study. Schizophrenia Research, 122, 232–238.

Chien

W. T.

Yeung

F. K. K.

Chan

A. H. L.

(2014). Perceived stigma of patients with severe mental illness in Hong Kong: Relationships with patients’ psychosocial conditions and attitudes of family caregivers and health professionals. Administration and Policy in Mental Health and Mental Health Services Research, 41, 237–251. doi:10.1007/s10488-012-0463-3

Chiu

M. Y. L.

W. W. N.

W. T. L.

Yiu

M. G.

(2010). Operationalization of the SAMHSA model of recovery: A quality of life perspective. Quality of Life Research, 19, 1–13. doi:10.1007/s11136-009-9555-2

Chung

K. F.

Wong

M. C.

(2004). Experience of stigma among Chinese mental health patients in Hong Kong. Psychiatric Bulletin, 28, 451–454.

Corrigan

P. W.

Salzer

Ralph

Sangster

Keck

(2004). Examining the factor structure of the Recovery Assessment Scale. Schizophrenia Bulletin, 30, 1035–1042.

Corrigan

P. W.

Watson

A. C.

Barr

(2006). The self-stigma of mental illness: Implications for self-esteem and self-efficacy. Journal of Social and Clinical, 25, 875–884.

10.

Davidson

Sells

Sangster

O’Connell

(2005). Qualitative studies of recovery: What we can learn from the person? In Ralph

R. O.

Corrigan

P. W.

(Eds.), Recovery in mental illness – Broadening our understanding of wellness (pp. 147–170). Washington, DC: American Psychological Association.

11.

Drapalski

A. L.

Lucksred

Perrin

P. B.

Aakre

J. M.

Brown

C. H.

DeForge

B. R.

… Boyd

J. E.

(2013). A model of internalized stigma and its effects on people with mental illness. Psychiatric Services, 64, 264–269.

12.

Evans-Lacko

Brohan

Mojtabai

Thornicroft

(2012). Association between public views of mental illness and self-stigma among individuals with mental illness in 14 European countries. Psychological Medicine, 42, 1741–1752.

13.

Fabrega

(1991). Psychiatric stigma in non-Western societies. Comprehensive Psychiatry, 32, 534–551.

14.

Fung

K. M.

Tsang

W. H.

Corrigan

P. W.

Lam

C. S.

Cheng

W. M.

(2007). Measuring self-stigma of mental illness in China and its implications for recovery. International Journal of Social Psychiatry, 53, 408–418.

15.

Gerlinger

Hauser

De Hert

Lacluyse

Wampers

Correll

C. U.

(2013). Personal stigma in schizophrenia spectrum disorders: A systematic review of prevalence rates, correlates, impact and interventions. World Psychiatry, 12, 155–164.

16.

Hsiao

F. H.

Klimidis

Minas

Tan

E. S.

(2005). Cultural attribution of mental health suffering in Chinese societies: The views of Chinese patients with mental illness and their caregivers. Journal of Clinical Nursing, 15, 998–1006. doi:10.1111/j.1365–2702.2006.01331.x

17.

Huck

S. W.

(2000). Reading statistics and research. New York, NY: Longman.

18.

IBM Corp. (2013). IBM SPSS Statistics for Windows (Version 22.0) [Computer software.]. Armonk, NY: Author.

19.

Lam

C. S.

Tsang

H. W. H.

Corrigan

P. W.

Lee

Y. T.

Beth

Kan

… Larson

J. E.

(2010). Chinese lay theory and mental illness stigma: Implications for research and practices. Journal of Rehabilitation, 76, 35–40.

20.

Lam

P. C.

Pan

Young

D. K.

(2015). Ways of coping of Chinese caregivers for family members with schizophrenia in two metropolitan cities: Guangzhou and Hong Kong, China. International Journal of Social Psychiatry, 61, 591–599. doi:10.1177/0020764014565797

21.

Leamy

Bird

Le Boutillier

Williams

Slade

(2011). Conceptual framework for personal recovery in mental health: Systematic review and narrative synthesis. British Journal of Psychiatry, 199, 445–452.

22.

Lee

Chiu

M. Y. L.

Tsang

Chui

Kleinman

(2006). Stigmatizing experience and structural discrimination associated with the treatment of schizophrenia in Hong Kong. Social Science & Medicine, 62, 1685–1696.

23.

Lee

M. T. Y.

Chiu

Kleinman

(2005). Experience of social stigma by people with schizophrenia in Hong Kong. British Journal of Psychiatry, 186, 153–157.

24.

Leung

K. F.

Wong

W. W.

Tay

M. S.

Chu

M. M.

S. S.

(2005). Development and validation of the interview version of the Hong Kong Chinese WHOQOL-BREF. Quality of Life Research, 14, 1413–1419.

25.

Leung

S. O.

Wong

P. M.

(2008). Validity and reliability of Chinese Rosenberg Self-Esteem Scale. New Horizons in Education, 56, 62–69.

26.

Livingston

J. D.

Boyd

J. E.

(2010). Correlates and consequences of internalized stigma for people living with mental illness: A systematic review and meta-analysis. Social Science & Medicine, 71, 2150–2161.

27.

Wolf

Wang

(2013). Experienced stigma and self-stigma in Chinese patients with schizophrenia. General Hospital Psychiatry, 35, 83–88. doi:10.1016/j.genhosppsych.2012.07.007

28.

Mak

W. W. S.

Cheung

R. Y. M.

(2012). Psychological distress and subjective burden of caregivers of people with mental illness: The role of affiliate stigma and face concern. Community Mental Health Journal, 48, 270–274. doi:10.1007/s10597-011-9422-9

29.

McNaught

Caputi

Oades

L. G.

Deane

F. P.

(2007). Testing the validity of the Recovery Assessment Scale using an Australian sample. Australian and New Zealand Journal of Psychiatry, 41, 450–457.

30.

R. M.

Pearson

Lam

Law

C. W.

Chiu

C. P.

Chen

E. Y.

(2008). What does recovery mean from schizophrenia mean? Perceptions of long-term patients. International Journal of Social Psychiatry, 54, 118–130.

31.

Philips

M. R.

Pearson

Yang

(2002). Stigma and expressed emotion: A study of people with schizophrenia and their family members in China. British Journal of Psychiatry, 181, 488–493.

32.

Ritsher

J. B.

Otilingam

P. G.

Grajiales

(2003). Internalized stigma of mental illness: Psychometric properties of a new measure. Psychiatry Research, 121, 31–49.

33.

Särndal

C. E.

Swensson

Wretman

(2003). Model assisted survey sampling. New York, NY: Springer.

34.

Schmitt

D. P.

Allik

(2005). Simultaneous administration of the Rosenberg self-esteem scale in 53 nations: Exploring the universal and culture-specific features of global self-esteem. Journal of Personality and Social Psychology, 89, 523–642.

35.

Sibitz

Provaznikova

Lipp

Lakeman

Amering

(2013). The impact of recovery-oriented day clinic treatment on internalized stigma: Preliminary report. Psychiatry Research, 209, 326–332.

36.

Skevington

S. M.

Lotfy

O’Connell

K. A.

(2004). The World Health Organization’s WHOQOL-BREF quality of life assessment: Psychometric properties and results of the international field trial – A report from the WHOQOL group. Quality of Life Research, 13, 299–310.

37.

Song

Hsu

(2011). The development of the Stages of Recovery Scale for persons with persistent mental illness. Research on Social Work Practice, 21, 572–581.

38.

Tang

I. C.

H. C.

(2012). Quality of life and self-stigma in Individuals with Schizophrenia. Psychiatric Quarterly, 83, 497–507.

39.

Tsang

H. W. H.

Angell

Corrigan

P. W.

Lee

Y. T.

Shi

Lam

C. S.

… Fung

K. M. T.

(2007). A cross sectional study of employers’ concerns about hiring people with psychiatric disorder: Implications for recovery. Social Psychiatry and Psychiatric Epidemiology, 42, 723–733.

40.

Tsang

H. W. H.

Tam

K. C.

Chan

Cheung

W. M.

(2003). Stigmatizing attitudes towards individuals with mental illness in Hong Kong: Implication for their recovery. Journal of Community Psychology, 31, 383–396.

41.

Tse

Ran

M. S.

Huang

Zhu

(2013). The urgency of now: Building a recovery-oriented, community mental health service in China. Psychiatric Services, 64, 613–616. doi:10.1176/appi.ps.000022013

42.

West

M. L.

Yanos

P. T.

Smith

S. M.

Roe

Lysaker

P. H.

(2011). Prevalence of internalized stigma among persons with severe mental illness. Stigma Research Action, 1, 3–10. doi:10.5463/sra.v1i1.9

43.

Xiang

Y. T.

Sartorius

Ungvari

G. S.

Chiu

H. F. K.

(2012). Mental health in China: Challenges and progress. The Lancet, 280, 1715–1716. doi:10.1016/S0140-6736(11)60893-3

44.

Yang

L. H.

(2007). Application of mental illness stigma theory to Chinese societies: Synthesis and new directions. Singapore Medical Journal, 48, 977–985.

45.

Young

K. W.

Pan

J. Y.

Cheng

(2015). Validity and reliability of Internalized Stigma of Mental Illness (Cantonese). Research on Social Work Practice. Advance online publication. doi:10.1177/1049731515576209