Destigmatized Group Intervention on Promoting Hope of Recovery: A Quasi-Experiment Study

Abstract

Objective:

This research project aims to evaluate the effectiveness of a destigmatized group intervention in promoting hope of recovery for adults with mental illness.

Methods:

A quasi-experimental research method was adopted. In addition to treatment as usual (TAU), the treatment group (n = 36) participated in a 10-session destigmatized group intervention, which was based on cognitive behavioral therapy (CBT), while the control group (n = 42) received TAU. Standardized assessment tools were used for outcome assessment at the pretreatment and posttreatment periods.

Results:

The 2 × 2 repeated measures analysis of covariance demonstrated that participants who participated in the destigmatized group intervention experienced significantly fewer self-stigma and more hope of recovery. Additionally, multiple linear regression showed that the reduction of self-stigma predicted the promotion of hope of recovery.

Conclusions:

This study supports the effectiveness of the destigmatized group intervention, suggesting the importance of reducing self-stigma for facilitating recovery for adults with mental illness.

Keywords

self-stigma hope of recovery depressive symptoms cognitive behavioral therapy destigmatized group intervention

Although various concepts and definitions of recovery have been proposed, to date there is no common consensus on its definition (Bellack, 2006; Cavelti, Kvrgic, Beck, Kossowsky, & Vauth, 2012; Corrigan & Ralph, 2005; Shanks et al, 2013). Generally, the definition and focus of recovery in the literature is in terms of either clinical or personal recovery (Cavelti et al., 2012). Clinical recovery focuses more on symptom alleviation and functional improvement, and various operational definitions of clinical recovery have been proposed (Rodgers, Norell, Roll, & Dyck, 2007). However, many researchers have also pointed out that from the perspective of community mental health service consumers, recovery should not be understood as an end product or a result but rather viewed as a unique personal journey or process (Anthony, 1993; Cavelti et al., 2012; Davidson, Sells, Sangster, & O’Connell, 2005; Drake & Whitley, 2014; Leamy, Bird, Le Boutillier, Williams, & Slade, 2011). Additionally, this approach has suggested essential components of personal recovery, such as hope; self-direction; empowerment; holistic, nonlinear, and strength-based approaches; peer support, responsibility; management of stigma; and a positive sense of self (Davidson et al., 2005; Drake & Whitley, 2014; Leamy et al., 2011). This study therefore adopts the approach of Anthony (1993), focuses on the recovery process, and defines personal recovery as a way of living a satisfying, hopeful, and contributing life, even with the limitations caused by mental illness.

Regaining hope of recovery has been regarded as a central component of personal recovery (Davidson et al., 2005; Leamy et al., 2011). Greater hope of recovery is found to be related to better recovery outcomes, such as better quality of life (Landeen, Pawlick, Woodside, Kirkpatrick, & Bryne, 2000), higher self-esteem (Mashiach-Eizenberg, Hasson-Ohayon, Yanos, Lysaker, & Roe, 2013; Young, Ng, Pan, Fung, & Cheng, 2017), fewer depressive symptoms (Schrank, Amering, Hay, Weber, & Sibitz, 2014), and fewer psychiatric symptoms (Waynor, Gao, Dolce, Haytas, & Reilly, 2012). Some psychosocial interventions have been suggested to foster hope of recovery, such as the Illness Management and Recovery Program (Salyers et al., 2009) and Wellness Recovery Action Planning (Cook et al., 2012).

Recovery is relevant to Chinese people with mental illness and plays a significant role in the lives of these individuals (Siu et al., 2012; Tse, Siu, & Kan, 2013; Young, Ng, & Pan, 2017; Young, Ng, Pan, Fung et al., 2017). However, as the research studies and service development on the recovery of people with mental illness in Hong Kong and other Chinese societies are at the beginning stage (Tse et al., 2013), little research has explored effective interventions for promoting the hope of recovery for people with mental illness in Chinese societies.

On the other hand, managing self-stigma properly and developing a positive sense of self has also been recommended as another important element of recovery for people with severe mental illness (Boyd, Adler, Otilingam, & Peters, 2014; Davidson et al., 2005). Research studies have suggested that it is feasible to reduce self-stigma in people with mental illness through short-term therapeutic groups (Lucksted et al., 2011; Mittal, Sulivan, Chekuri, Allee, & Corrigan, 2012). In particular, short-term cognitive behavioral therapy (CBT) groups have been shown to be effective in reducing self-stigma of people with mental illness (Lucksted et al., 2011; Macinnes & Lewis, 2008; Shimotsu et al., 2014; Young, 2018). Additionally, research evidence also suggests that self-stigma is negatively related to personal recovery (Gerlinger et al., 2013; Livingston & Boyd, 2010).

According to the cognitive model of self-stigma proposed by Corrigan and Rao (2012) and Corrigan, Watson, and Barr (2006), self-stigmatizing beliefs are regarded as irrational beliefs held by the person with mental illness, which lead to his or her depressive mood and reduced hope of recovery. CBT and related intervention techniques have been shown to be effective in challenging and reducing these irrational, self-stigmatizing beliefs (Lucksted et al., 2011; Macinnes & Lewis, 2008; Shimotsu et al., 2014; Young, 2018).

In view of the effectiveness of CBT on reducing self-stigma, a destigmatized group intervention, which was based on CBT, has been developed by the first author and shown to be effective on reducing self-stigma for people with mental illness (Young, 2018). In particular, this destigmatized group intervention differed from the traditional CBT by putting more emphasis on developing positive beliefs in one’s self and recognizing one’s own strengths (e.g., “I can achieve my own goals even I have mental illness”; “I can take care of myself and do household duties properly, and so I am not useless nor being family burden”) through cognitive intervention techniques such as guided discovery (Ng, Tsun, Su, & Young, 2013). Additionally, by using techniques of behavioral experiments and homework assignments, participants learned to recognize their strengths and started to make contributions to the family, such as doing specific household duties, and/or to serve the community-based mental health service center through short-term, simple volunteer services. In this way, by helping participants to develop positive beliefs in one’s self and recognize one’s own strengths, the destigmatized group intervention promoted hope of recovery for participants. This idea is supported by a previous study, which reported that CBT could help people with mental illness to develop positive beliefs in themselves and reduce hopelessness (Hodgekins & Fowler, 2010). Moreover, as empowerment has been advocated by researchers to reduce self-stigma (Brohan, Elgie, Sartorius, & Thornicroft, 2010; Lucksted et al., 2011), the destigmatized group intervention was also different from traditional CBT by emphasizing empowering participants to reduce self-stigma through skills learning as well as receiving emotional support among participants. Participants shared their current specific and individualized stigmatized social situations during the first two group sessions and then they learned to deal with these individualized stigmatized social situations through social skill training and role-play within group sessions as well as practicing these newly learned skills in their real-life situations. On the other hand, by sharing similar difficulties encountered with public stigma, participants learned to normalize their experiences of public stigma. By receiving emotional support as well as sharing of resources and skills among group members for responding to public stigma, participants could reduce their feelings of loneliness, helplessness, and hopelessness and enhance their self-efficacy to face public stigma. In this way, by emphasizing empowering participants, the destigmatized group intervention can help people with mental illness face social stigma and promote their hope of recovery. However, whether the reduction of self-stigma promotes the hope of recovery remains unclear, as very few research studies have been done in this area.

Aims

This study aims to evaluate the effectiveness of the destigmatized group intervention to reduce self-stigma and promote hope of recovery for people with mental illness in a Chinese society. Additionally, this study also aims to examine the direct effect of the reduction of self-stigma on the promotion of hope of recovery for people with mental illness.

Hypotheses

It is hypothesized that:

Hypothesis 1: Participants who participated in the destigmatized group intervention will experience significantly more reduction in self-stigma than those participants who received treatment as usual (TAU).

Hypothesis 2: Participants who participated in the destigmatized group intervention will experience significantly more reduction in depressive symptoms than those participants who received TAU.

Hypothesis 3: Participants who participated in the destigmatized group intervention will experience significantly higher level of hope of recovery than those participants who received TAU.

Hypothesis 4: The reduction of self-stigma due to participation in the destigmatized group intervention predicts the reduction of depressive symptoms of participants.

Hypothesis 5: The reduction of self-stigma due to the participation in the destigmatized group intervention predicts the promotion of hope of recovery of participants.

Method

In this study, a quasi-experimental design was adopted. Subjects were recruited from local four Integrated Community Centres for Mental Wellness (ICCMWs) run by nongovernmental organizations (NGOs) through open recruitment from ICCMWs’ promotional activities and receiving referrals from social workers of the ICCMWs. Each ICCMW provided services for over 1,000 people with severe mental illness in Hong Kong. A cohort of participants who provided their consent to participate in the research project were assigned to the treatment group. Another cohort of participants, whose characteristics were similar to the treatment group participants, were recruited as the control group after they provided consent to participate in the research project. Standardized assessment tools were used for data collection at the ICCMWs at the pre and posttreatment periods by a research assistant. Each participant received a coupon with amount of HK$100 as an incentive payment after completing all assessments. Data collection was started in April 2015 and completed in August 2016.

Client Inclusion Criteria

In this study, the inclusion criteria were participants who (a) were 18 years or older, (b) had received a diagnosis of a mental disorder from a medical officer according to the Diagnostic and Statistical Manual of Mental Disorder, fifth edition (American Psychiatric Association, 2013), (c) were receiving community-based mental services from a local NGOs, and (d) had given consent to this research work.

Sample Size Consideration

Due to constraints in resources, only four intervention groups could be conducted.

Treatment Group

The destigmatized group intervention involved in this study was based on CBT and included the following group components, which were thought to be essential elements for reducing self-stigma: (1) altering stigmatized beliefs by using cognitive restructuring techniques, (2) developing positive beliefs in one’s self and recognizing one’s own strengths, (3) social skill training on responding to public stigmatizing situations, and (4) enhancing social support among participants (Corrigan et al., 2006; Fung, Tsang, & Cheung, 2011; Lucksted et al., 2011; Mittal et al., 2012; Shimotsu et al., 2014; Young, 2018). A standardized and detailed program manual was designed and written for the treatment group, which is available from the first author on request.

In addition to the TAU provided by the local ICCMW, the treatment group participants attended a 10-session destigmatized group held at multiple sites in Hong Kong. Each session was conducted once per week and, on average, was 90 min in length. A registered and experience social worker who possessed a master’s in social work and had more than 2 years of experience working with people with mental illness was employed by this project to conduct the destigmatized group. The social worker was given the program manual and received training and supervision from the first author. Additionally, the social worker used group work skills to facilitate social support among treatment group participants.

Control Group

The control group participants received TAU, that is, usual services provided by an ICCMW, such as interest classes and leisure activities.

Outcome Assessment Tools

The primary outcomes are the reduction of self-stigma and improvement of hope of recovery. The secondary outcome is the reduction of depressive symptoms.

The Recovery Assessment Scale (RAS) has been considered by researchers to be one of the most reliable and valid measures for assessing personal recovery and is widely used internationally (Cavelti et al., 2012; Chiba, Miyamoto, & Kawakami, 2010; Corrigan, Salzer, Ralph, Sangster, & Keck, 2004; Law, Morrison, Byrne, & Hodson, 2012; MacNaught, Caputi, Oades, & Deane, 2007; Salzer & Brusilovskiy, 2014). The construct validity of the Chinese version of RAS has been confirmed, with strong internal consistency (α = .92; Young, Ng, Pan, Fung, et al., 2017). In this study, the 9-item Hope subscale of RAS (RAS-H) was used to assess the hope of recovery for participants. Each item was rated on a 5-point scale ranging from 1 (strongly disagree) to 5 (strongly agree). The scores were summed over the items, such that higher scores indicated better hope of recovery. In this study, the RAS-H had very good internal consistency (α = .87).

The Internalized Stigma of Mental Illness (ISMI) has been validated for the assessment of self-stigma for people with mental illness (Ritsher, Otilingam, & Grajiales, 2003) and has been widely used internationally (Boyd et al., 2014). The 24-item ISMI is used to measure self-stigma; it consists of four subscales: Shame, Stereotype Endorsement, Perceived Discrimination, and Social Withdrawal. Each item was rated on a 4-point scale ranging from 1 (strongly disagree) to 4 (strongly agree). Item scores were averaged, with higher scores indicating greater self-stigma. The construct validity of the Chinese version of ISMI (24 items) has been confirmed, with good internal consistency (α = .93; Young, Ng, Pan, & Cheng, 2017). As suggested by previous studies (Brohan et al., 2010; Ritsher et al., 2003; Young, Ng, Pan et al., 2017), the cutoff point on the ISMI is ≥2.5, which was adopted in the present study. In this study, the ISMI showed excellent internal consistency (α = .93).

The Beck Depression Inventory (BDI) is a widely used assessment scale to measure self-reported depression (Beck, Steer, & Brown, 1996). The BDI has 21 items, and each item was rated on a 4-point scale ranging from 0 to 3. The scores were summed over the items such that higher scores indicated more severity in depressive symptoms. The reliability and validity of the BDI (Chinese version) has been tested and shown to be satisfactory (α = .91; Byrne, Stewart, & Lee, 2004). The BDI used in this project is a 21-item questionnaire with each item rated on a 4-point Likert-type scale that ranges from 0 (symptom not present) to 3 (symptom strongly present). As suggested by previous studies (Beck et al., 1996), a score of ≥14 is considered to be the cutoff point for clinical depression, which was adopted in the present study. In this study, the internal consistency of BDI was excellent (α = .94).

Data Analysis

The pretreatment demographic and clinical characteristics between the treatment and control group were compared by using χ² tests for the categorical variables and one-way analysis of variance (ANOVA) for continuous variables. For Hypotheses 1–3, the differences in the changes in outcome measures between the treatment and control group were analyzed using general linear model 2 (group) × 2 (time) repeated measures analysis of covariance (ANCOVA), with the two groups (treatment and control group) as between-subject factors and the two outcome assessments (pre and posttreatment scores) as within-subject factors, adjusting for group differences in various demographic variables, including age, gender, marital status, education, source of income, diagnosis, period of illness, number of hospitalization, centers for implementing the treatment groups, and pretreatment outcome assessment score. The effect sizes were calculated using the partial η². For partial η², the values of 0.01, 0.06, and 0.14 were considered small, moderate, and large effects, respectively (Cohen, 1988).

With regard to Hypotheses 4 and 5, data analyses were limited to those treatment group participants who completed the destigmatized group intervention. The identification of demographic and clinical variables that were related to the improvements in the outcome assessment scores (i.e., ΔISMI, ΔBDI, and ΔRAS-H) among the treatment group was conducted by using a one-way ANOVA for the categorical variables and Spearman’s correlation analysis for the continuous variables. In identifying the predicting variables for the ΔBDI and ΔRAS-H scores, multiple linear regression analyses were conducted with the ΔBDI and ΔRAS-H scores included as the dependent variables and all the demographic and clinical variables were the independent variables. A stepwise selection method was adopted to explore and compare the importance of the individual independent variables in predicting the ΔBDI and ΔRAS-H scores. The collinearity of each regression model was examined with a variance inflation factor (VIF); a VIF less than 10 was considered noncollinear. Statistical analyses were performed using SPSS 25.0 (IBM Corporation, 2017). For all analyses, two-tailed p values < .05 indicated statistical significance.

Ethical Considerations

The ethical considerations of this study were evaluated and approved by the Research Committee of the Hong Kong Baptist University (Ref. no. FGR2/14-15/019). Written informed consent was obtained from all participants on the day of pretreatment assessment.

Research Results

This study involved 112 adults with mental illness. Thirty-four who did not meet the inclusion criteria were excluded from this study. Thirty-six participants were assigned to the treatment group, while 42 participants were assigned to the control group. In addition to TAU provided by the ICCMW, the treatment group participated in a 10-session destigmatized group intervention, while the control group received TAU provided by the ICCMW, including interest classes and leisure activities. Four treatment groups and four control groups were formed. Each treatment group consisted of 6–10 participants with mental illness. Among treatment group participants, 4 dropped out and 32 completed the destigmatized group intervention, with an overall attendance rate of 88.8%. Two participants who completed the destigmatized group intervention were lost to follow up and did not complete the post-assessment, while one control participant did not complete the post-assessment. The seven individuals who failed to complete the assessment resulted in an attrition rate of 9.0%. Figure 1 illustrates the flow of participants through each stage of the study.

Figure 1.

Flow of participants through each stage of the study. Adapted from: Des Jarlais, D. C., Lyles, C., Crepaz, N., and the Trend Group. (2004). For more information, visit: http://www.cdc.gov/trendstatement/

Characteristics of Clients

Analyses from the t test and χ² test showed no significant differences between the completers (n = 71) and noncompleters (n = 7) for all of the pretreatment demographic and clinical variables. Additionally, the latter did not give any prior notice or reason for their withdrawal from the study, which is perhaps not uncommon for people with mental illness in Hong Kong. Thus, the missing data could be assumed to be missing at random. To maintain the full sample for data analysis, a multiple imputation method was conducted to replace the missing data prior to the data analysis.

As shown in Table 1, analyses from the t test and χ² test showed no significant differences between the treatment and control group for all except one pretreatment demographic variable. The treatment group had a higher percentage of participants relying on their own salary and family financial support for their living expenses compared with the control group.

Table 1.

Demographic and Clinical Characteristics of Participants.

Characteristics	Treatment Group	Control Group	Total	χ²/t (p)
Characteristics	(N = 36)	(N = 42)	(N = 78)	χ²/t (p)
Female, n (%)^a	33 (91.67)	32 (76.19)	65 (83.33)	3.343 (.067)
Age, mean ± SD (years)^b	50.33 ± 10.53	45.98 ± 10.38	47.99 ± 10.61	−1.836 (.066)
Marital status, n (%)^a				4.977 (.083)
Single	17 (47.22)	18 (42.86)	35 (44.87)
Married	15 (41.67)	11 (26.19)	26 (33.33)
Divorce or separated	4 (11.11)	13 (30.95)	17 (21.80)
Widowed
Education level, n (%)^a				0.582 (.901)
Post-secondary	8 (22.22)	9 (21.43)	17 (21.79)
Secondary	19 (52.77)	24 (57.14)	43 (55.13)
Primary	7 (19.44)	8 (19.05)	15 (19.23)
None	2 (5.56)	1 (2.38)	3 (3.85)
Occupational status, n (%)^a				5.269 (.153)
OE (FT)	3 (8.33)	3 (7.14)	6 (7.69)
OE (PT)	4 (11.11)	0 (0.00)	4 (5.13)
SW/SE	2 (5.56)	4 (9.52)	6 (7.69)
Unemployed, housewife, retired	27 (75.00)	35 (83.33)	62 (79.49)
Main source of income, n (%)^a				7.920 (.048)*
Salary/saving	7 (19.44)	3 (7.14)	10 (12.82)
Family support	17 (47.22)	14 (33.33)	31 (39.74)
Governmental support	11 (30.56)	25 (59.52)	36 (46.16)
Other	1 (2.78)	0 (0.00)	1 (1.28)
Place of residence, n (%)^a				5.128 (.163)
Public housing	15 (41.67)	25 (59.52)	40 (51.28)
Private housing	20 (55.56)	14 (33.33)	34 (43.59)
Half-way house	0 (0.00)	2 (4.76)	2 (2.56)
Other	1 (2.78)	1 (2.38)	2 (2.56)
Living situation, n (%)^a				3.070 (.215)
With family	28 (80.00)	26 (65.00)	54 (72.00)
Alone	7 (20.00)	12 (30.00)	19 (25.33)
Others	0 (0.00)	2 (5.00)	2 (2.67)
Diagnosis^a				1.570 (.814)
Schizophrenia	10 (27.78)	13 (30.96)	23 (29.48)
Depression	18 (50.00)	21 (50.00)	40 (51.28)
Anxiety	1 (2.78)	3 (7.14)	4 (5.13)
Bipolar affective disorder	6 (16.66)	4 (9.52)	9 (11.53)
Others	1 (2.78)	1 (2.38)	2 (2.56)
Period of illness, mean ± SD ^b	15.00 ± 10.59	12.39 ± 9.55	13.57 ± 10.05	−1.127 (.260)
Number of hospitalizations, mean ± SD ^b	1.74 ± 3.18	1.68 ± 2.45	1.71 ± 2.79	−0.078 (.938)

^aPearson χ². ^bIndependent sample t test.

*Significant at p < .05.

Taking all participants together, most of the participants were female (83.3%, n = 65), had attained an education to the secondary school level (76.9%, n = 60), were living with their family members (72.0%, n = 54), and had a mean age of 48.0 years (SD = 10.6). Nearly half (44.9%, n = 35) were single, and one third (33.3%, n = 26) were married, while the rest were divorced, widowed, or separated. Half were living in public housing (51.2%, n = 40), two fifths were living in a privately owned flat (43.6%, n = 34), while only a few (2.6%, n = 2) were living in psychiatric halfway houses. The majority of participants (79.5%, n = 62) were unemployed, housewives, or retired, while only one eighth (12.7%, n = 10) were engaged in full-time or part-time open employment. Half of the participants were diagnosed with depression (51.3%, n = 40), followed by schizophrenia (29.5%, n = 23), bipolar (11.5%, n = 9), and anxiety disorders (5.1%, n = 4). Three quarters (77.3%, n = 61) of participants had suffered from mental illness for 5 years or more, with a mean of 13.6 years (SD = 10.1). Half (51.3%, n = 39) of the participants had been admitted into mental hospitalization one or more times, with a mean of 1.7 times (SD = 2.8).

Pretreatment Assessment Scores

As shown in Table 2, the results of ANOVA showed no significant difference between the treatment and control group in all pretreatment outcome assessment scores, including RAS-H, BDI, ISMI, and ISMI subscales.

Table 2.

Pretest, Posttest, and Comparison of Change Score on Self-Stigma and Other Assessment Scores Between the Treatment and Control Group.

	Treatment Group (n = 36)		Control Group (n = 42)		Between Group Analysis
	Baseline Score(Mean ± SD)	Change Score(Mean ± SD)	Baseline Score(Mean ± SD)	Change Score(Mean ± SD)	t (p)^a	F (p)^b	Partial η²
ISMI	2.52 ± 0.59	−0.36 ± 0.51	2.41 ± 0.56	−0.03 ± 0.42	−0.812 (.417)	7.482 (.008)**	.099
ISMI-Sh	2.77 ± 0.72	−0.39 ± 0.59	2.63 ± 0.70	0.06 ± 0.54	−0.859 (.390)	7.799 (.007)**	.103
ISMI-Se	2.19 ± 0.53	−0.26 ± 0.41	2.19 ± 0.54	−0.05 ± 0.47	−.037 (.970)	0.836 (.364)	.012
ISMI-Pd	2.52 ± 0.74	−0.32 ± 0.78	2.42 ± 0.68	0.07 ± 0.70	−0.578 (.563)	4.225 (.043)*	.059
ISMI-Sw	2.64 ± 0.76	−0.47 ± 0.74	2.43 ± 0.69	0.07 ± 0.57	−1.231 (.218)	8.754 (.004)**	.114
BDI	23.28 ± 15.08	−7.33 ± 11.57	24.64 ± 16.44	−1.83 ± 7.17	0.379 (.305)	4.362 (.040)*	.060
RAS-H	30.44 ± 6.57	2.97 ± 4.31	29.45 ± 6.90	0.05 ± 4.22	0.647 (.517)	4.698 (.034)*	.065

Note. ISMI = Internalized Stigma of Mental Illness Scale Self-Stigma subscale; ISMI-Sh = ISMI Shame subscale; ISMI-Se = ISMI Stereotype Endorsement subscale; ISMI-Pd = ISMI Perceived Discrimination subscale; IMIS-Sw = ISMI Social Withdrawal subscale; BDI = Back Depression Inventory; RAS-H = Recovery Assessment Scale Hope subscale.

^aIndependent sample t test for baseline outcome assessment scores between the treatment and control groups. ^b2 × 2 repeated measures of analysis of covariance adjusting for age, gender, education level, main source of income, period of illness, number of hospitalization, and baseline assessment score.

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

Taking all participants together, half (51.3%, n = 40) of participants reported self-stigma as assessed by ISMI (i.e., ISMI ≥ 2.5). They reported a mean ISMI score of 2.46 (SD = .57). Among the four subscales of the ISMI, participants gave shame/alienation the highest ranking (M = 2.70, SD = .70), followed by social withdrawal (M = 2.53, SD = .70) and perceived discrimination (M = 2.47, SD = .70), while stereotype endorsement was ranked the lowest (M = 2.19, SD = .53). Additionally, two thirds (69.2%, n = 54) of participants were found to have depressive mood as assessed by BDI (i.e., BDI score ≥ 14). They reported a mean ISMI score of 24.01 (SD = 15.74). Additionally, participants reported a moderate level of hope of recovery, as assessed by RAS-H (M = 29.9, SD = 6.72).

Moreover, the ISMI score was significantly negatively related to RAS-H score (Spearman’s ρ = −.48, p < .01) and positively related to BDI score (ρ = −.63, p < .01).

Treatment Outcomes

Hypothesis 1: Reduction of Self-Stigma

Table 2 summarizes the change of outcome measures of both treatment and control group. The results of the 2 × 2 repeated measures of ANOVA indicated that after controlling for differences between the treatment and control group in almost all demographic variables, including age, gender, marital status, education, source of income, diagnosis, number of hospitalization, period of illness, centers for implementing the treatment groups, and pretreatment outcome assessment score, the destigmatized group intervention was significantly more effective than the control group in reducing the ISMI score (F = 7.48, p < .01), with a moderate to large effect size (partial η² = .10). This finding indicates participants who participated in the destigmatized group intervention experienced significantly more reduction in self-stigma than those participants who received TAU.

Hypothesis 2: Reduction of Depressive Symptoms

The results of the 2 × 2 repeated measures of ANOVA indicated that after controlling for differences between the treatment and control group in almost all demographic variables, including age, gender, marital status, education, source of income, diagnosis, number of hospitalization, period of illness, centers for implementing the treatment groups, and pretreatment outcome assessment score, the destigmatized group intervention was significantly more effective than the control group in reducing the BDI score (F = 4.36, p < .05), with a moderate effect size (partial η² = .06; see Table 2). This finding indicates participants who participated in the destigmatized group intervention experienced significantly more reduction in depressive symptoms than those participants who received TAU.

Hypothesis 3: Promotion of Hope of Recovery

The results of the 2 × 2 repeated measures of ANOVA indicated that after controlling for differences between the treatment and control group in almost all demographic variables, including age, gender, marital status, education, source of income, diagnosis, number of hospitalization, period of illness, centers for implementing the treatment groups, and pretreatment outcome assessment score, the destigmatized group intervention was significantly more effective than the control group in improving the RAS-H score (F = 4.70, p < .05), with a moderate effect size (partial η² = .07; see Table 2). This finding indicates participants who participated in the destigmatized group intervention experienced significantly better hope of recovery than those participants who received TAU.

Hypothesis 4: Reduction of Self-Stigma Predicting Reduction of Depressive Symptoms

As shown in Table 3, the results of χ² and Spearman’s correlation analysis showed that the ΔBDI score was not related to all demographic variables but was related to the pretreatment BDI score (ρ = −.57, p < .01), pretreatment ISMI score (ρ = −.42, p < .05), ΔISMI score (ρ = .60, p < .01), and ΔRAS-H score (ρ = .40, p < .05).

Table 3.

Variables Related to the Improvement of Self-Stigma Scores for Treatment Group Participants.

	ΔISMI		ΔBDI		ΔRAS-H
	F/ρ	p Value	F/ρ	p Value	F/ρ	p Value
Sex^a	0.010	.922	0.695	.410	0.033	.858
Diagnosis^a	0.914	.445	1.008	.402	0.613	.611
Marital status^a	1.129	.336	0.035	.966	2.319	.114
Education level^a	1.225	.317	0.190	.903	0.745	.533
Occupational status^a	1.028	.393	0.260	.854	0.866	.469
Source of income^a	0.796	.505	1.899	.150	0.196	.899
Place of residence^a	1.923	.162	2.207	.126	0.372	.692
Living situation^a	3.657	.065	0.325	.573	0.024	.877
Age^b	−0.133	.450	−0.195	.264	0.051	.777
Period of illness^b	−0.190	.271	0.204	.240	0.085	.626
Number of hospitalization^b	0.103	.555	0.311	.070	−0.101	.563
ISMI^b	−0.607	.000**	-0.421	.010*	0.579	.000**
RAS-H^b	0.565	.000**	0.261	.130	−0.742	.000**
BDI^b	−0.286	.093	−0.569	.000**	0.272	.117
ΔISMI^b	1.000	—	0.597	.000**	−0.779	.000**
ΔRAS-H^b	−0.779	.000**	0.397	.016*	1.000	—
ΔBDI^b	0.597	.000**	1.000	—	−0.397	.016*

Note. ISMI = Internalized Stigma of Mental Illness Scale; RAS-H = Recovery Assessment Scale Hope subscale; BDI = Back Depression Inventory; ΔISMI = change of ISMI score from baseline to posttreatment; ΔBDI = change of BDI score from baseline to posttreatment; ΔRAS-H = change of RAS-H score from baseline to posttreatment; ρ = Spearman’s rho.

^aAnalysis of covariance. ^b Spearman correlation.

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

On the other hand, Table 4 shows the final stepwise multiple linear regression model of ΔBDI score for treatment group. Collinearity was verified, and the collinearity statistics suggested that collinearity could be neglected for all of the regression models (VIF < 10). The final regression model for ΔBDI explained 58.2% of its total variance. The pretreatment BDI was the strongest predictor of ΔBDI and contributed to one third (33.8%) of its total variance, while the ΔISMI and pretreatment RAS-H scores contributed to 17.5% and 6.9% of the total variance of ΔBDI, respectively.

Table 4.

Stepwise Multiple Linear Regression Analysis on ΔRAS-H and ΔBDI for the Treatment Group.

Model	Predictor	Adj. R²	B	β	t	F	Collinearity Statistics
Model	Predictor	Adj. R²	B	β	t	F	Tolerance	VIF
Dependent variable: ΔRAS-H
1	(Constant)	.608	−.574		−.806	55.265**
	ΔISMI		−8.179	−.787	−7.434**		1.000	1.000
2	(Constant)	.738	10.968		3.939**	50.413**
	ΔISMI		−6.198	−.596	−6.120**		.787	1.270
	RAS-H		−.355	−.413	−4.239**		.787	1.270
Dependent variable: ΔBDI
3	(Constant)	.338	4.658		1.426	18.844**
	BDI		−.514	−.597	−4.341**		1.000	1.000
4	(Constant)	.513	5.786		2.053*	19.400**
	ΔISMI		10.696	.450	3.633**		.906	1.104
	BDI		−.395	−.459	−3.702**		.906	1.104
5	(Constant)	.582	30.165		3.043**	17.256**
	ΔISMI		13.503	.569	4.593**		.779	1.284
	BDI		−.510	−.593	−4.698**		.749	1.336
	RAS-H		−.679	−.345	−2.549*		.650	1.537

Note. ISMI = Internalized Stigma of Mental Illness Scale Self-stigma subscale; BDI = Back Depression Inventory; RAS-H= Recovery Assessment Scale Hope subscale; ΔISMI = change in ISMI score from baseline to posttreatment; ΔRAS-H= change of RAS-H score from baseline to posttreatment; Adj. R ² = adjusted R ²; ΔR ² = change of adj. R ².

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

Hypothesis 5: Reduction of Self-Stigma Predicting Improvement of Hope of Recovery

As shown in Table 3, the results of χ² and Spearman’s correlation analysis showed that the ΔRAS-H score was not related to all demographic variables but was related to the pretreatment RAS-H score (ρ = −.74, p < .01), pretreatment ISMI score (ρ= .58, p < .05), ΔISMI score (ρ = −.78, p < .01), and ΔBDI score (ρ = −.40, p < .05).

On the other hand, Table 4 shows the final stepwise multiple linear regression model of ΔRAS-H score for the treatment group. Collinearity was verified, and the collinearity statistics suggested that collinearity could be neglected for all of the regression models (VIF < 10). The final regression model for ΔRAS-H explained about three quarters (73.8%) of its total variance. The ΔISMI was the strongest predictor of ΔRAS-H and contributed to more than two thirds (60.8%) of its total variance, while the pretreatment RAS-H score contributed to 13.0% of the total variance of ΔRAS-H.

Discussion and Application to Practice

In this study, the demographic characteristics of participants largely resemble those of a recent, larger-scale, local, cross-sectional survey that involved a sample of 266 participants randomly recruited from a population of people with severe mental illness who were receiving community-based mental health services provided by different NGOs in Hong Kong (Young & Ng, 2015). In this study, half (51.3%) of participants reported having self-stigma (i.e., IMSI score ≥ 2.5) at pretreatment, which was higher than the 36.1% reported in the United States (West, Yanos, Smith, Roe, & Lysaker, 2011), 41.7% reported in Europe (Brohan et al., 2010), 38.3% in Hong Kong, and 49.5% in Guangzhou, China (Young & Ng, 2015). The higher prevalence of self-stigma among participants in this study was likely because those participants who joined in the indigenized destigmatizing groups were those who already had problems with self-stigma and who intended to handle this problem by joining in and learning from the treatment group. Also, as self-stigma is positively related to depressive symptoms and negatively related to hope of recovery (Gerlinger et al., 2013; Livingston & Boyd, 2010), the high prevalence rate of self-stigma of participants may contribute to the relatively high prevalence (i.e., 69.2%) of pretreatment clinical depression (BDI score ≥ 14) and a relatively lower level of pretreatment hope of recovery of participants.

Consistent with Hypothesis 1, the results of the 2 × 2 repeated measures of ANCOVA demonstrated that participants who participated in the destigmatized group intervention experienced significantly more reduction in self-stigma than those participants who received TAU. According to the cognitive model of self-stigma (Corrigan et al., 2006; Corrigan & Rao, 2012), self-stigmatizing beliefs are regarded as irrational beliefs upheld by the person with mental illness and thus CBT has been recommended to help these individuals challenge these irrational self-stigmatizing beliefs (Fung et al., 2011; Lucksted et al., 2011). As shown in this study, participation in the destigmatized group intervention, which is mainly based on CBT and related intervention techniques, is effective in helping participants dispute irrational self-stigmatizing beliefs (e.g., “I cannot recover from mental illness”; “I am useless and become family burden”), which is consistent with previous research studies (Macinnes & Lewis, 2008; Mittal et al., 2012; Shimotsu et al., 2014; Young, 2018).

Hypotheses 2 and 4 are supported. In particular, the results of the 2 × 2 repeated measures of ANCOVA demonstrated that participants who participated in the destigmatized group intervention experienced significantly more reduction in depressive symptoms than those participants who received TAU. Additionally, the results of multiple linear regression demonstrated that the reduction of self-stigma due to the completion of the destigmatized group intervention predicted or led to the reduction of depressive symptoms. It has been suggested that the irrational self-stigmatizing beliefs negatively impact individuals’ lives by increasing depressive symptoms, and thus after challenging these irrational self-stigmatized beliefs by CBT, the depressive symptoms of people with mental illness would be reduced accordingly (Corrigan et al., 2006; Shimotsu et al., 2014). The above notions are supported by the results of this study. As shown by the above research results, after challenging and reducing the irrational self-stigmatizing beliefs in the destigmatized group intervention through the use of CBT and related intervention skills, the depressive symptoms of participants were reduced accordingly.

Hypotheses 3 and 5 are supported. In particular, the results of the 2 × 2 repeated measures of ANCOVA demonstrated that participants who participated in the destigmatized group intervention experienced significantly better hope of recovery than those participants who received TAU. Additionally, the results of multiple linear regression demonstrated that the reduction of self-stigma due to the completion of the destigmatized group intervention predicted or led to the promotion of hope of recovery. As self-stigma is found to be negatively related to personal recovery (Boyd et al., 2014), it is reasonable to assume that the irrational self-stigmatizing beliefs negatively impact individuals’ lives by inducing a sense of hopelessness and reducing the hope of recovery, so after challenging these irrational self-stigmatizing beliefs by CBT and related intervention skills, the hope of recovery of people with mental illness increases accordingly. The above notions are supported by the research results of this study. As shown by the above research results, after challenging and reducing the irrational self-stigmatizing beliefs in the destigmatized group intervention through the use of CBT and related intervention skills, the hope of recovery of participants improved accordingly.

This present study provides evidence to support the efficacy and effectiveness of a destigmatized group intervention based on CBT in promoting hope of recovery and reducing depressive symptoms for people with mental illness. Additionally, this study is one of the few studies to demonstrate that the reduction of stigma predicts/leads to the promotion of hope of recovery and reduction of depressive symptoms of people with mental illness. The above research findings of this study have important implications for community-based mental health services by suggesting that reducing self-stigma can foster hope of recovery and reducing depressive symptoms. Thus, it is recommended for recovery-orientated services to implement the short-term destigmatized group intervention that can be easily implemented by an experienced social worker to facilitate recovery for people with mental illness.

Several methodological limitations of this quasi-experiment study require attention. First, the generalizability of the research results is limited by the nonrandomized research design and small sample size. Second, the attrition rate of 9.0% reported in this study may give rise to a biased estimation of the intervention effects. Nevertheless, the similarities between completers and noncompleters of this study in almost all of the pretreatment characteristics contribute to minimizing any bias in the research results. In the future, it would be better to conduct a larger-scale, randomized controlled study to further validate the effectiveness of the destigmatized group intervention model on reducing self-stigma and improving hope of recovery for people with mental illness.

Footnotes

Acknowledgments

This study was generously fully sponsored by the Hong Kong Baptist University [Ref. No: FRG2/14-15/019].

Declaration of Conflicting Interests

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

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Daniel K. W. Young

References

American Psychiatric Association. (2013). Diagnostic and Statistical Manual of Mental Disorders (5th ed.). 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. doi:10.1037/h0095655

Beck

A. T.

Steer

R. A.

Brown

G. K.

(1996). Manual for the Beck Depression Inventory-II. San Antonio, TX: Psychological Corporation.

Bellack

A. S.

(2006). Scientific and consumer models of recovery in schizophrenia: Concordance, contrast, and implications. Schizophrenia Bulletin, 32, 432–442.

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. doi:10.1016/j.schres.2010.02.1065

Byrne

B. M.

Stewart

Lee

P. W. H.

(2004). Validating the beck depression inventory-II for Hong Kong community adolescents. International Journal of Testing, 4, 199–216. doi:10.1207/s15327574ijt0403_1

Cavelti

Kvrgic

Beck

E. M.

Kossowsky

Vauth

(2012). Assessing recovery from schizophrenia as an individual process. A review of self-report instruments. European Psychiatry, 27, 19–32. doi:10.1016/j.eurpsy.2011.01.007

Chiba

Miyamoto

Kawakami

(2010). Reliability and validity of the Japanese version of Recovery Assessment Scale (RAS) for people with chronic mental illness: Scale development. International Journal of Nursing Studies, 47, 314–322. doi:/10.1016/j.ijnurstu.2009.07.006

10.

Cohen

(1988). Statistical power analysis for the behavioral sciences (2nd ed.). New Jersey, Hillsdale: Lawrence Erlbaum.

11.

Cook

J. A.

Copeland

M. E.

Jonikas

J. A.

Hamilton

M. M.

Razzano

L. A.

Grey

D. D.

… Boyd

(2012). Results of a randomized controlled trial of mental illness self-management using Wellness Recovery Action Planning. Schizophrenia Bulletin, 38, 881–891.

12.

Corrigan

P. W.

Ralph

R. O.

(2005). Recovery as consumer vision and research paradigm. In Ralph

R. O.

Corrigan

P. W.

(Eds.), Recovery in mental illness—Broadening our understanding of wellness (pp. 3–18). Washington, DC: American Psychological Association.

13.

Corrigan

P. W.

Rao

D. P.

(2012). The impact of self-stigma and mutual help program on quality of life of people with serious mental illness. Community Mental Health Journal, 49, 1–6.

14.

Corrigan

P. W.

Salzer

Ralph

Sangster

Keck

(2004). Examining the factor structure of the Recovery Assessment Scale. Schizophrenia Bulletin, 30, 1035–1042. doi:10.1093/oxfordjournals.schbul.a007118

15.

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 Psychology, 25, 875–884. doi:10.1521/jscp.2006.25.8.875

16.

Davidson

Sells

Sangster

O’Connell

(2005). Qualitative studies of personal 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.

17.

Des Jarlais

D. C.

Lyles

Crepaz

, and the Trend Group. (2004). Improving the reporting quality of nonrandomized evaluations of behavioral and public health interventions: The TREND statement. American Journal of Public Health, 94, 361–366.

18.

Drake

R. E.

Whitley

(2014). Recovery and severe mental illness: Description and analysis. Canadian Journal of Psychiatry, 59, 236–242.

19.

Fung

K. M. T.

Tsang

H. W. H.

Cheung

W. M.

(2011). Randomized controlled trial of the self-stigma reduction program among individuals with schizophrenia. Psychiatric Research, 189, 208–214. doi:10.1016/j.psychres.2011.02.013

20.

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. doi:10.1002/wps.20040

21.

Hodgekins

Fowler

(2010). CBT and recovery from psychosis in ISREP trial: Mediating effects of hope and positive beliefs on activity. Psychiatric Services, 61, 321–324.

22.

IBM Corporation. (2017). IBM SPSS Statistics for Windows (Version 25.0) . Armonk, NY: Author.

23.

Landeen

Pawlick

Woodside

Kirpatrick

Byrne

(2000). Hope, quality of life, and symptom severity in individuals with schizophrenia. Psychiatric Rehabilitation Journal, 23, 364–369. doi:10.1037/h0095142

24.

Law

Morrison

Byrne

Hodson

(2012). Recovery from psychosis: A user informed review of self-report instruments for measuring recovery. Journal of Mental Health, 21, 193–208. doi:10.3109/09638237.2012.670885

25.

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.

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 and Medicine, 71, 2150–2161.

27.

Lucksted

Drapalski

Calmes

Forbes

DeForge

Boyd

(2011). Ending self-stigma: Pilot evaluation of a new intervention to reduce internalized stigma among people with mental illness. Psychiatric Rehabilitation Journal, 35, 51–54. doi:10.2975/35.1.2011.51.54

28.

Macinnes

D. L.

Lewis

(2008). The evaluation of a short group programme to reduce self-stigma in people with serious and enduring mental health problems. Journal of Psychiatric and Mental Health Nursing, 15, 59–65. doi:10.1111/j.1365-2850.2007.01212.x

29.

Mashiach-Eizenberg

Hasson-Ohayon

Yanos

Lysaker

P. H.

Roe

(2013). Internalized stigma and quality of life among persons with severe mental illness: The mediating roles of self-esteem and hope. Psychiatric Research, 208, 15–20.

30.

MacNaught

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.

31.

Mittal

Sulivan

Chekuri

Allee

Corrigan

P. W.

(2012). Empirical studies of self-stigma reduction strategies: A critical review of literature. Psychiatric Services, 63, 974–981.

32.

Tsun

Young

K. W.

(2013). Cognitive behavioural therapy in the Chinese cultural context: A case report. Asia-Pacific Psychiatry, 5, 205–211. doi:10.1111/appy.12092

33.

Ritsher

J. B.

Otilingam

P. G.

Grajiales

(2003). Internalized stigma of mental illness: Psychometric properties of a new measure. Psychiatry Research, 121, 31–49. doi:10.1016/j.psychres.2003.08.008

34.

Rodgers

M. L.

Norell

D. M.

Roll

J. M.

Dyck

D. G.

(2007). An overview of mental health recovery. Primary Psychiatry, 14, 76–85. Retrieved from http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.623.9165&rep=rep1&type=pdf

35.

Salyers

M. P.

Godfrey

J. L.

McGuire

A. B.

Gearhart

Rollins

A. L.

Boyle

(2009). Implementing the illness management and recovery program for consumers with severe mental illness. Psychiatric Services, 60, 483–490.

36.

Salzer

Brusilovskiy

(2014). Advancing recovery science: Reliability and validity properties of the Recovery Assessment Scale. Psychiatric Services, 65, 442–453.

37.

Schrank

Amering

Hay

A. G.

Weber

Sibitz

(2014). Insight, positive and negative symptoms, hope, depression and self-stigma: A comprehensive model of mutual influences in schizophrenia spectrum disorders. Epidemiology and Psychiatric Services, 23, 271–279. doi:10.1017/S2045796013000322

38.

Shanks

Williams

Leamy

Bird

V. J.

Le Boutillier

Slade

(2013). Measures of personal recovery: A systematic review. Psychiatric Services, 64, 974–980.

39.

Shimotsu

Horikwa

Emura

Ishikawa

S. I.

Nagao

Ogata

… Hosomi

(2014). Effectiveness of group cognitive behavioural therapy in reducing self-stigma in Japanese psychiatric patients. Asian Journal of Psychiatry, 10, 39–44. doi:10.1016/j.ajp.2014.02.006

40.

Siu

B. W. M.

B. F. L.

V. C. K.

Yeung

Y. M.

Lee

M. K. L.

Leung

A. Y. H.

(2012). Mental health recovery for psychiatric inpatient services: Perceived importance of the elements of recovery. East Asian Archives of Psychiatry, 22, 39–48. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/22714873

41.

Tse

Siu

W. M.

Kan

A. L.

(2013). Can recovery-oriented mental health services be created in Hong Kong? Struggles and strategies. Administration and Policy in Mental Health and Mental Health Services Research, 40, 155–158. doi:10.1007/ s10488-011-0391-7

42.

Waynor

W. R.

Gao

Dolce

J. N.

Haytas

L. A.

Reilly

(2012). The relationship between hope and symptoms. Psychiatric Rehabilitation Journal, 35, 345–348. doi:10.2975/35.4.2012.345.348

43.

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

44.

Young

K. W.

(2018). Cognitive behavioral therapeutic group in reducing self-stigma for people with mental illness. Research on Social Work Practice, 28, 827–837.

45.

Young

K. W.

(2015). The prevalence and predictors of self-stigma of individuals with mental health illness in two Chinese cities. International Journal of Social Psychiatry, 62, 176–185.

46.

Young

K. W.

Pan

J. Y.

(2017). Predictors and prevalence of recovery and remission for consumers discharged from mental hospitals in a Chinese society. Psychiatric Quarterly, 88, 839–851.

47.

Young

K. W.

Pan

J. Y.

Cheng

(2017). Validity and reliability of Internalized Stigma of Mental Illness (Cantonese). Research on Social Work Practice, 27, 103–110.

48.

Young

K. W.

Pan

J. Y.

Fung

Cheng

(2017). Validity and reliability of recovery assessment scale for Cantonese speaking Chinese consumers with mental Illness. International Journal of Mental Health and Addiction, 15, 198–208.