Schizophrenia Research 173 (2016) 45–53
Contents lists available at ScienceDirect
j ourna l homepage: www.e lsev ie r .com/ locate /schres
Therapeutic intervention for internalized stigma of severemental illness: A systematic review and meta-analysis
Hector W.H. Tsang ⁎, S.C. Ching, K.H. Tang, H.T. Lam, Peggy Y.Y. Law, C.N. Wan Department of Rehabilitation Sciences, The Hong Kong Polytechnic University, Hong Kong
⁎ Corresponding author at: Neuropsychiatric Rehabilita Rehabilitation Sciences, the Hong Kong Polytechnic Un Hong Kong.
E-mail address: email@example.com (H.W.H. Ts
http://dx.doi.org/10.1016/j.schres.2016.02.013 0920-9964/© 2016 Published by Elsevier B.V.
a b s t r a c t
a r t i c l e i n f o
Article history: Received 3 August 2015 Received in revised form 5 February 2016 Accepted 5 February 2016 Available online 9 March 2016
Objective: Internalized stigma can lead to pervasive negative effects among people with severe mental illness (SMI). Although prevalence of internalized stigma is high, there is a dearth of interventions and meanwhile a lack of evidence as to their effectiveness. This study aims at unraveling the existence of different therapeutic in- terventions and the effectiveness internalized stigma reduction in people with SMI via a systematic review and meta-analysis. Methods: Five electronic databases were searched. Studies were included if they (1) involved community or hos- pital based interventions on internalized stigma, (2) included participants who were given a diagnosis of SMI N 50%, and (3) were empirical and quantitative in nature. Results: Fourteen articles were selected for extensive review and five for meta-analysis. Nine studies showed sig- nificant decrease in internalized stigma and two showed sustainable effects. Meta-analysis showed that there was a small to moderate significant effect in therapeutic interventions (SMD = −0.43; p = 0.003). Among the intervention elements, four studies suggested a favorable effect of psychoeducation. Meta-analysis showed that there was small to moderate significant effect (SMD = −0.40; p = 0.001). Conclusion: Most internalized stigma reduction programs appear to be effective. This systematic review cannot make any recommendation on which intervention is more effective although psychoeducation seems most promising.More Randomized Controlled Trials (RCT) on particular intervention components using standard out- come measures are recommended in future studies.
© 2016 Published by Elsevier B.V.
Keywords: Psychoeducation Severe mental illness Self-stigma Internalized stigma Psychosocial intervention
Internalized stigma, also called self-stigma, is theprocess of adopting the public’s stigmatizing opinions into own thoughts. Its prevalence among people with severe mental illness (SMI) stands at 41.7% (Gerlinger et al., 2013). Meta-analysis (Livingston and Boyd, 2010) found that high levels of internalized stigmawere significantly correlat- edwith hopelessness, poorer self-esteem, reduced empowerment/mas- tery, and reduced self-efficacy. Internalized stigma significantly predicted poor social functioning over time (Fung et al., 2008; Tsang et al., 2010; Yanos et al., 2012a, 2012b). Higher internalized stigma was also associated with poorer quality of life on all the domains of WHOQOL-Brief (Mosanya et al., 2014). In addition, higher levels of in- ternalized stigma were significantly related to more severe psychiatric symptoms, poorer treatment adherence, and lower utilization ofmental health services (Rüsch et al., 2009). Up to 20% of people may even dis- continue treatment prematurely due to internalized stigma (Corrigan,
tion Laboratory, Department of iversity, Hung Hom, Kowloon,
2014). Unfortunately, poorer treatment adherence was related to poorer treatment outcomes, more re-hospitalization, and increased health costs (Lacro et al., 2002). All of the above studies suggest that in- ternalized stigma has tremendous negative impact on functional out- comes of people with SMI and increases societal burden in taking care of these individuals.
Given the close negative relationship between internalized stigma and recovery (Yanos et al., 2008), interventions that attempt to reduce it among those with SMI so as to improve their prospect for recovery are important both in psychiatric rehabilitation and community integration.
Although a conceptual framework (Hayward and Bright, 1997) on internalized stigma reduction is available, there is a dearth of interven- tions and meanwhile a lack of evidence as to their effectiveness. The programs usually consist of a combination of different elements of inter- vention such as psychoeducation, cognitive behavioral therapy, and so- cial skills training. Currently there are no widely accepted treatment protocols on internalized stigma reduction. Numerous studies on inter- nalized stigma were published in the past few years. There are a few reviews (e.g., Mittal et al., 2012) on this area. But no attempt on meta- analysis has ever been conducted. To fill the knowledge gap, this systematic review and meta-analytical study focuses on studies which
46 H.W.H. Tsang et al. / Schizophrenia Research 173 (2016) 45–53
are randomized control trials, clinical trials, or experimental trials in na- ture and reported therapeutic interventions to reduce internalized stig- ma in people with SMI.
The PRISMA Statement Criteria (Liberati et al., 2009) was adopted throughout the review to report our results.
2.1. Data sources
Systematic searchwas done to locate studies from PubMed, PsychInfo (1806-Present), SCI (1970-Present), SSCI (1970-Present), and Scopus on October 3rd, 2014. The search terms were generated based on the four types of eligibility criteria mentioned above. They are presented in Supplementary Appendix 1. In addition, emails were sent to a few prominent researchers on mental illness stigma that had close collabo- rations with the first author. One replied and suggested an in press arti- cle that we missed in the systematic search and eventually included in the systematic review.
2.2. Study selection
Studieswere included if the following criteria weremet: (1) studies: randomized clinical trials, clinical trials, and experimental studies studying internalized stigma reduction in people with SMI which was operationally defined as mental illness having a chronic course and leading to significant social and occupational dysfunction such as Schizophrenia, Psychotic disorder, Psychosis, Delusional disorder, Schizoaffective, Bipolar disorder and Personality disorder; (2) partici- pants: more than 50% studies participants were given a diagnosis of SMI,; (3) intervention: studies that compared community or hospital based therapeutic interventions with conventional treatment; (4) out- come measures: studies that used validated instruments for screening and assessing the severity of internalized stigma. Qualitative studies and literature reviews were excluded.
There were no limitations in the follow up period. Search was re- stricted to publications in English but there was no limit to the years of publication.
Two reviewers independently conducted the search in five electron- ic databases. The results were screened on titles and abstracts by all re- viewers independently. Disagreements were resolved via discussion between reviewers with the facilitation of the corresponding author.
2.3. Data extraction and analysis
Data extracted from selected studies included: (1) characteristics of the study (e.g., aims, study design, intervention setting, participants’ characteristics, randomization procedures, inclusion and exclusion criteria, etc.); (2) outcome measures (e.g., outcome measure instru- ments, assessment period, dropout rate, etc.); (3) interventions (e.g., intervention approach, trainers’ quality, etc.); and (4) study results (e.g., significant effects, sustainability, etc.). To ensure accuracy, five re- viewers assessed two to three studies independently and the data were cross-checked by another reviewer. Also, emails were sent to the authors for clarifying missing or unclear data.
2.4. Assessment of methodological quality
Themethodological risk of bias for each trial was evaluated using the risk of bias table (Higgins and Green, 2011). Five reviewers worked in- dependently to determine the adequacy of randomization sequence, blinding of patients and assessors, and the extent of follow up. Disagree- ments were resolved in discussion.
2.5. Data synthesis and analysis
Review Manager 5.3, developed by the Cochrane Collaboration (2014), was used for meta-analysis of the results of included studies. The outcomes across the trials were expressed in Cohen’s D. Standard- ized mean differences (SMDs) were calculated for the pooled effects. SMDs were interpreted using the “rule of thumb”: 0.2 represents a small effect, 0.5 represents a moderate effect, and 0.8 represents a large effect (Cohen, 1988). Heterogeneity was tested with an X2 test. I2
was also reported. I2 statistic N75% was considered to have high degree of heterogeneity while I2 statistic of 25%–50% was considered to have a low degree of heterogeneity (Higgins et al., 2003). We used random- effects model for heterogeneity (p b 0.05) and a fixed effects model for heterogeneity (p N 0.05). Sensitivity analyses were conducted with psychoeducation group and professional-led or peer-led intervention. Publication bias was examined using funnel plot. A value of p b 0.05 was considered statistically significant.
Since the outcome of internalized stigma reduction program was assessed by more than one tool in a trial, only the primary outcome (i.e., internalized stigma reduction) was included in the meta-analysis. Emails were sent to corresponding authors for clarifying missing data.
3.1. Results of literature search
Eight hundred nineteen articles were retrieved. Three hundred eighty duplicated and 421 irrelevant articles were excluded after initial screening of title and abstract. Full reports of 22 studies were acquired and eight were further excluded for the following reasons: (1) not a clinical trial, (2) b50% participants given diagnoses of SMI, (3) explor- atory research, and (4) duplicated studies (Fig. 1). (See Figs. 2– 4.)
3.2. Description of included studies
Fourteen studies, including seven RCTs (Corrigan et al., 2015; Çuhadar and Çam, 2014; Fung et al., 2011; McCay et al., 2007; Russinova et al., 2014; Rüsch et al., 2014; Yanos et al., 2012a, 2012b), three controlled clinical trials (Roe et al., 2014; Sibitz et al., 2013; Uchino et al., 2012), and four uncontrolled studies without a control group (Costain et al., 2014; Lucksted et al., 2011; Lysaker et al., 2012; Staring et al., 2013)met inclusion criteria. These studieswere originated from nine countries across Americas, Europe, and Asia. Most programs adopted psychoeducation approach with inclusion of a combination of other components such as CBT, social skills training, goal attainment program, and narrative therapy. The duration of these programs ranged from 10 to 40 sessions (Russinova et al., 2014; Sibitz et al., 2013). Other than the above more conventional therapeutic elements, there were two evolving innovative interventions. Both of them were peer led group interventions. One was Coming Out Proud (Corrigan et al., 2015; Rüsch et al., 2014). It consisted of group discussion focusing on topics of secrecy and disclosure of own mental illness. The other was Photovoice (Russinova et al., 2014). In the group, individuals photo- graph objects or events in their daily lives were used to generate narra- tives for group discussion. The characteristics of included studies are summarized in Table 1.
Participants were given a diagnosis of schizophrenia, schizophrenia spectrum disorder, bipolar disorder, or major mood disorder. One study included patients with schizophrenia only (Fung et al., 2011) and one included patients with bipolar disorder only (Çuhadar and Çam, 2014). Participants were mostly given a diagnosis by psychiatrists according to DSM IV or ICD-10 (Yanos et al., 2012a, 2012b). Two studies recruited participants according to their self-reported diagnosis. Only one study reported the use of the structured interviewprocedure in ver- ifying the diagnosis of the research participants (Lysaker et al., 2012).
Fig. 1. Selection process of included studies.
47H.W.H. Tsang et al. / Schizophrenia Research 173 (2016) 45–53
Sample size of the studies varied from 21 to 205 participants, with a total of 1131 participants including 879participants in the experimental groups and 452 participants in the control groups.
Controlled group design was employed in ten studies. Treatment as usual (TAU)was applied in seven studies (McCay et al., 2007; Roe et al., 2014; Russinova et al., 2014; Rüsch et al., 2014; Sibitz et al., 2013; Uchino et al., 2012; Yanos et al., 2012a, 2012b), conventional treatment group (i.e., newspaper reading) in one study (Fung et al., 2011), no ac- tive treatment in one study (Çuhadar and Çam, 2014), and waitlist con- trol in one study (Corrigan et al., 2015). All studies provided information
Fig. 2. A forest plot of the meta-analysis of three studies comp
on the duration of treatment in patient intervention groups. Thirteen in- cluded studies examined the effect of internalized stigma reduction and one examined the effect of negative symptoms immediately following the internalized stigma reduction intervention. In addition, seven stud- ies (Corrigan et al., 2015; Costain et al., 2014; Fung et al., 2011; Lysaker et al., 2012; Russinova et al., 2014; Rüsch et al., 2014; Yanos et al., 2012a, 2012b) examined the sustainability of intervention effect after follow- up periods ranging from three weeks to six months.
Regarding outcome measures, internalized stigma was assessed as the primary outcome in all the included studies except one which
aring psychoeducation to TAU for changes in ISMI score.
Fig. 3. A forest plot of the meta-analysis of three studies comparing self-stigma reduction program to TAU for changes in ISMI total score.
48 H.W.H. Tsang et al. / Schizophrenia Research 173 (2016) 45–53
treated internalized stigma as a mediator of dysfunctional beliefs (Staring et al., 2013). The internalized stigma scales applied in the stud- ies included Internalized Stigma of Mental Illness (ISMI) (Costain et al., 2014; Çuhadar and Çam, 2014; Lucksted et al., 2011; Lysaker et al., 2012; Roe et al., 2014; Russinova et al., 2014; Rüsch et al., 2014; Sibitz et al., 2013; Yanos et al., 2012a, 2012b), short form of Self-stigma of Mental Illness Scale (SSMIS) (Corrigan et al., 2015), Chinese Self- stigma of Mental Illness Scale (CSSMIS) (Fung et al., 2011), Link Per- ceived Stigma Questionnaire (LPSQ) (McCay et al., 2007), and Japanese version of Social Distance Scale (SDS-J) (Uchino et al., 2012). Other outcomemeasurements were also applied to assess the interven- tion effect, for examples, the Positive and Negative Syndrome Scale (PANSS) and the Rosenberg Self-Esteem Scale (RSES). These outcome measures varied substantially across studies.
3.3. Results of individual studies
Nine studies reported significant internalized stigma reduction. Studies that reported significant effects of therapeutic intervention other than internalized stigma reduction are shown in Table 1. Seven studies had follow-up periods but only two studies showed sustainable effects for one to three months (Corrigan et al., 2015; Russinova et al., 2014).
3.4. Effects of intervention approaches on internalized stigma
Meta-analysis was performed among studies which used similar treatment approaches (e.g., psychoeducation, CBT, etc.) and similar out- comemeasures. Following this, six controlled studies were identified to use ISMI as outcomemeasure. However, one of the studies was eventu- ally excluded (Çuhadar and Çam, 2014) because it used the Turkish ver- sion of ISMI which was not comparable to the original version. Consequently, five trials were included in meta-analysis (Roe et al., 2014; Russinova et al., 2014; Rüsch et al., 2014; Sibitz et al., 2013; Yanos et al., 2012a, 2012b).
3.4.1. Comparison 1: psychoeducation element Four out of seven trials which contained psychoeducation were
RCTs. Two trials indicated significant effects when compared with con- trolled group with no active treatment (Çuhadar and Çam, 2014) and TAU (Russinova et al., 2014). The former focused on patients with bipo- lar disorder and the latter focused on patients with schizophrenia
Fig. 4. A forest plot of the meta-analysis of four studies comparing self-stigma reduct
spectrum disorder and bipolar disorder. ISMI total scores from two RCTs and one controlled trial were pooled and the pooled SMD was −0.40 [−0.64, −0.16], indicating a small-moderate effect (p = 0.001). There was a low degree of heterogeneity (I2 = 17%).
3.4.2. Comparison 2: cognitive behavioral therapy element Two out of five trials containing CBT element were RCTs. Both trials
suggested no intergroup differencewhen compared to newspaper read- ing group (Fung et al., 2011) and TAU (Yanos et al., 2012a, 2012b). The former focused on patientswith schizophrenia and the latter focused on patients with schizophrenia, schizoaffective, and bipolar disorder.
3.4.3. Comparison 3: group discussion element Three out of the four trialswhichhad groupdiscussion elementwere
RCTs. These RCTs included innovative interventions, namely photovoice (Russinova et al., 2014) and Coming Out Proud (COP) (Corrigan et al., 2015; Russinova et al., 2014). Photovoice was a discussion group on the selected photographs regarding to stigma while COP was a peer- led group intervention which empowered people with mental illness to decide on secrecy and disclosure of illness. When compared to TAU, the trial on photovoice indicated significant effect (Russinova et al., 2014). For COP, one of them suggested significant effect on reduction in internalized stigma when compared to waitlist control (Corrigan et al., 2015). Another trial, however, suggested no group x time effect (Rüsch et al., 2014). All of them focused on patients with seriousmental illness.
3.4.4. Comparison 4: social skills training element Oneout of two trials that contained social skills training elementwas
RCT. It suggested no intergroup difference on patients with schizophre- nia when compared to newspaper reading group (Fung et al., 2011).
3.4.5. Comparison 5: narrative enhancement and cognitive therapy (NECT) element
This approach focused on the process of telling and sharing personal narratives about their experience of mental illness. Two out of three tri- als containing NECT element were RCTs. When compared to TAU, one trial indicated significant effect (Russinova et al., 2014) while another suggested no intergroup difference (Yanos et al., 2012a, 2012b). Both focused on patients with schizophrenia spectrum disorder and bipolar disorder. Moreover, the former focused on depressive disorder.
ion program to TAU for changes in average of the sum of sub-scales 1–4 of ISMI.
49H.W.H. Tsang et al. / Schizophrenia Research 173 (2016) 45–53
3.4.6. Other intervention approaches There was one RCT which reported the effect of group psychothera-
py on internalized stigma reduction among individuals with schizo- phrenia spectrum disorder (McCay et al., 2007). No intergroup difference was reported. The other two trials were uncontrolled clinical trials involving genetic counseling (Costain et al., 2014) and vocational rehabilitation (Lysaker et al., 2012).
Five studies were eligible for meta-analysis (Roe et al., 2014; Russinova et al., 2014; Rüsch et al., 2014; Sibitz et al., 2013; Yanos et al., 2012a, 2012b). ISMI included 5 subscales: alienation, stereotype endorsement, discrimination experience, social withdrawal, and stigma resistance. The total scores from three psychosocial interventions were pooled and the pooled SMDwas−0.43 [−0.72,−0.14]which indicated a small-moderate effect (p = 0.003) of psychosocial interventions. There was a low degree of heterogeneity (I2 =22%). As two studies ex- cluded the 5th subscale (i.e., stigma resistance) of ISMI score (Roe et al., 2014; Rüsch et al., 2014), we pooled the ISMI scores by taking average of the sum of sub-scales 1–4 from four studies. Similar to the result of total ISMI score, it indicated a small to moderate effect (SMD = −0.33 [−0.54, −0.12]; p = 0.002). The results from both professional led and peer led psychosocial intervention were pooled. Result of meta- analysis showed moderate significant effect professional led interven- tion (SMD = −0.51 [−0.79, −0.23]; p b 0.001) but not peer led intervention.
Quality assessment for each trial is presented in Table 2. None of the studies met all of the criteria for high quality. Inclusion and exclusion criteria were clearly defined except one trial (Roe et al., 2014). Seven studies were RCTs with randomization method reported clearly except one study (Corrigan et al., 2015). Common problems included inade- quate blinding of patients and outcome assessors, lack of information on sample size estimation, and intention-to-treat analyses not applied.
This paper presents the most updated systematic review and meta- analysis on internalized stigma reduction. Half of the reviewed studies were published from 2013 to 2015 which were not included in earlier reviews. The studies were conducted in nine countries across America, Europe, andAsiawhich suggest cross-cultural some degree of generaliz- ability of the results.
As pointed out in the introduction, programs to reduce internalized stigma consist of many components. Our review reveals the beneficial effects of the psychoeducation component in internalized stigma reduc- tion when compared to TAU. For COP and NECT, effects were inconsis- tent. For CBT and social skills training, RCTs reported no significant effects.
Interpretation and generalization of results should be cautiously treated due to limited number andmethodological problems of studies. Because of bias among studies, the results of meta-analysis should also be read with caution. The studies used different diagnostic criteria for participants, including self-reported and diagnoses given by psychia- trists. Moreover, the proportion of participants with SMI was slightly less than 50% in the control group in one of the studies (Rüsch et al., 2014). Both of these factors limit the comparability between results for the participants with SMI in the studies. The baseline internalized stigma level among participants varied, with mean total scores of ISMI at baseline ranging from 1.63 to 2.23 out of 4.0. Although there does not have any evidence, this variation in baseline level of internalized stigmamight affect the effectiveness of programs and hence the compa- rability of effectiveness between different studies. Some of the studies only presented group × time effect or main effect. As intergroup differ- ence was not presented in all studies, the significance of between group differences was unclear. Randomization was involved only in seven
studies. Twelve trials did not adhere to the intention-to-treat principle. These two factors lead to possible over-estimation of the treatment ef- fect. There is also a possibility of publication bias. Among the limited number of studies, different combinations of intervention approaches and outcome measurements were adopted. It is therefore hard to draw an accurate conclusion based on our meta-analysis. Finally, as few trials involved post-intervention follow up or the follow-up period was short, it is difficult to make conclusion on the sustainability of the interventions.
To date, a variety of interventions are available but there is a lack of commonly accepted framework on internalized stigma reduction. The framework proposed by Hayward and Bright (1997) adopted CBT as the main intervention with the role of psychosocial factors and self- management skills also taken into consideration.
There are some implications for further research. Our meta-analysis shows that psychoeducation is common among the studies and seems to have significant effect. However, among these three studies, only two (Roe et al., 2014; Russinova et al., 2014)mentioned clearly the con- tent of psychoeducation, which included understanding of stigma and strategies to cope with stigma. Given that psychoeducation was not the sole intervention in these studies and the follow-up was short, it is difficult to make a solid recommendation as to the optimal duration of psychoeducation program. It is however suggested that future studies should focus on studying the effectiveness of psychoeducation as the only intervention with longer follow up period to evaluate the sustain- ability of its outcome. As the content of psychoeducation varies, it is also necessary to study the specific content that actually contributes to its ef- fectiveness. Our systematic review found that there were only limited studies meeting our inclusion criteria. Most studies had numerous lim- itations in measurement tools and methodological qualities. These added to our difficulties in making conclusion based on our results. Fol- lowing this, we suggest the use of common outcome measures among studies such as ISMI, and improved methodological qualities of future studies to facilitate similar reviews in the future.
Two evolving innovative interventions, Coming Out Proud and Photovoice, deserve more attention. The results from trials of COP were inconclusive. Yet the study on Photovoice had a significant effect in reduction of internalized stigma and was shown sustainable in a pe- riod of three months. As new and promising interventions, more re- search on their effectiveness is needed. As both of these two innovative interventions involve peer-led activities, the therapeutic values of this type of activities should be more carefully discerned.
Other than research, this study has some implications for clinical practice. First, the serious consequences of leaving internalized stigma untreated should be brought to the awareness of clinicians. Second, in- ternalized stigma reduction programs have to be delivered by health and rehabilitation professionals. Relevant training courses should be made available to those professionals who show interest in putting these programs into their clinical and rehabilitation practice. Third, if re- courses are limited, practitioners should be recommended to select those programs with a psychoeducation element because it is found to be the most evidence-based in reducing internalized stigma based on existing evidence. Finally, clinicians should be encouraged to explore more innovative approaches to counteract internalized stigma because existing interventions are limited and their evidence is not entirely compelling.
Most programs we reviewed showed significant effects in reducing internalized stigma. As an emerging area, studies were still limited. The programs are implemented by professionals including clinicians, psychiatric nurse, social worker, and non-professionals who are trained according to the program manuals. Among different intervention ap- proaches, psychoeducation seems to be more promising and two novel techniques (i.e., coming out proud and photovoice) deserve
Table 1 Summary of the effects of self-stigma reduction program.
Study Participants’ diagnosis
Intervention Outcomes Result
Design; Country Inclusion Sample Size (Pre−/Post)
Approach (frequency) Control Duration of follow-up
Corrigan et al., 2015
RCT; U.S. People with serious mental illness who have life disabilities
EG: 107/51 CG: 98/75
Coming Out Proud (COP) (2 h session, 3 sessions, over 3 days or 1 day long group)
Waitlist 1 month 1. (Short form) SSMIS 2. Stigma Stress Scale 3. CES-D
Effect in subscales under SSMIS: 1. Group × time: no significant effect in change in ‘harm” comparing COP group to control group. Significant reduction in “harm” from pre to posttest (p b 0.01) and pre to follow-up for COP group. (p b 0.05) 2. Group × time: significant interaction for change in applying stereotypes to self across groups (F(2,98) = 3.74, p b .05). Significant reduction in applying stereotypes to self from pre to posttest (p b 0.05) and at one month follow-up (p b 0.05) 3. Group × time: significant interaction for agreeing on stereotype 4. Group × time: no significant change in “awareness” Stigma Stress Scale 1. Group × time: significant in stigma-related harm 2. Group × time: no significant for resources to cope with stigma Depression 1. Group × gender × time: significant difference in CES-D
Costain et al., 2014
Uncontrolled clinical trial; Canada
Schizophrenia Schizoaffective disorder
25/22 (21 for FU)
Genetic counseling (1 session with mean 46.4 [SD 11.1] minutes)
NA 7 weeks 1. ISMI 1. Decrease in self-stigma (p = 0.0111) 2. Reduction in self-blame (p = 0.0401) 3. Improvement in understanding of the empiric recurrence risk (p = 0.0090) 4. Reduction in associated concern (p = 0.0020) 5. Significant increase in subjective (p = 0.0007) and objective (p = 0.0103) knowledge
Çuhadar and Çam, 2014
RCT; Turkey Bipolar disorder 63/47 Psychoeducation program (90 min session; once a week; 7 sessions)
NA 1. ISMI (Turkish Version) 2. BDFQ
1. Significant decrease in total ISMI score (p b 0.001) and sub-scale 1–4 (p b 0.001) 2. Significant gain in domestic relationship subscales (p = 0.03)
Fung et al., 2011
RCT; Hong Kong Schizophrenia CG: 32/30 EG: 34/34
Self-stigma reduction program (twice a week, 6 weeks; new life: once 1 week, 12 weeks)
3 months 1.CSSMIS 2.CAQ-SPMI 3.PTCS 4.BPRS 5.GAF 6.SUMD 7.CGSS
1. Group × time: significant improvement in the self-esteem decrement subscale of CSSMIS *others subscale: stereotype agreement and self-concurrence show trend of improvement 2. Group × time: significant changes in stages of change continuous score of CAQ-SPMI and the participation subscale of PTCS 3. Group × time: no significant changes found in SUMD and CGSS
Lucksted et al., 2011 
Pilot study; U.S. Self reported: Schizophrenia Schizoaffective disorder Major mood disorder
EG: 50/34 Ending self-stigma (ESS) with lectures and group discussions. (90 min session; once a week, 9 weeks)
NA NA 1. ISMI 2. MHRM 3. MSPSS 4. Empowerment Scale (10 out of 28 items)
1. Group × time: significance in self stigma reduction.(p b 0.01) 2. Group × time: recovery orientation (MHRM) and social support (MSPSS) significantly increase 3. Group × time: mean empowerment increased but insignificantly (p = 0.0556 N 0.05) 4. Group × time: all ISMI subscale scores increased significantly except stigma resistance
Lysaker et al., 2012
Schizophrenia Schizoaffective disorder
EG: 70/70 Vocational rehabilitation: part-time paid work placement (non-specified frequency)
NA 5 months 1. ISMI (5th scale excluded) 2. MSEI 3. PANSS
1. Group × time: modest but significant decrease in internalized stigma (p = 0.007) 2. Group effect: significant for both emotional discomfort and self-esteem 3. Time effect: significant for emotional discomfort and self-esteem 4. Group × time: significant for self-esteem 5. No group or time effect for positive symptoms
McCay et al., 2007
RCT; Canada Schizophrenia Schizophreniform disorder Schizoaffective disorder
EG: 67/47 Psychotherapy group (1.5 h session; once a week;12 weeks)
TAU NA 1.MES 2.TSCS 3.RSES 4.SES 5.LPSQ 6.QLS 7.MHS 8.PANSS 9.GAF
1. Group × time: no significance improvement in self concept, self esteem, self efficacy, and stigma 2. Group × time: significant changes on measure of engulfment, hope and quality of life measures compared with control
50 H .W
Russinova et al., 2014
RCT; US Schizophrenia spectrum disorder Bipolar disorder Depressive disorder Other diagnosis
EG: 40/38 CG: 42/37
Antistigma Photovoice (90 min per session, 10 weeks)
TAU 3 months 1. ISMI 2. ACSS 3. PGRS 4. Empowerment scale 5. CES-D 6. GPSES
1. Significant decrease in overall ISMI score (p = 0.03) in experimental group 2. Significant more likely to report using proactive strategies: challenging (p = 0.04), educating others (p = 0.01) to cope with perceived stigma. 3. Significantly greater increases in community activism and autonomy (p = 0.02) on empowerment scale 4. Significant increases in overall level of PRGS (p = 0.04)
Roe et al., 2014
Controlled clinical trials; Israel
SMI ≥ 50% EG: 222/119 NECT (1-h session; once a week; 20 sessions)
TAU NA 1. ISMI 2. MANSA 3. ADHS 4. RSE
1. A significant group × time interaction (F(4114) = 2.68, p b 0.04) and with higher improvement in self-stigma in the treatment group. 2. A significant group × time interaction (p = 0.043) and with higher improvement in quality of life in the treatment group. 3. No significant group × time effect 4. A significant group × time interaction (p = 0.003) and with higher improvement in self-esteem in the treatment group.
Rüsch et al., 2014
Pilot RCT; Switzerland
Self reported: Schizophrenia spectrum disorder Bipolar disorder Depressive disorder
CG: 50/47 EG:50/40
Coming Out Proud (COP): with peer-led group discussion. (2 h session, once a week, 3 sessions)
TAU 3 weeks 1. ISMI (5th scale excluded) 2. Empowerment scale 3. Link’s 5-item secrecy scale 4. COMIS (7 items) 5. CAS (8 item) 6. Stress appraisal score 7. CES-D (15-item German version)
1. Group × time: no significant effect detected on self-stigma as primary outcome. 2. Group × time: no significant effect of treatment on empowerment or on disclosure-related self-efficacy. 3. Group × time: significant pre-post effect in some secondary outcome including decreasing stigma stress and secrecy as compared to the control group, with a medium-effect size for stigma stress reduction 4. The effect is only partly sustainable. Disclosure-related distress decreased significantly in the COP group from baseline to post-intervention and follow up. Perceived benefits of being out increased at post-intervention and remained stable for follow up.
Sibitz et al., 2013
Controlled clinical trial; Austria
Schizophrenia, Schizoaffective disorder, other F2 disorders
EG: 46/40 CG: 51/40
Recovery oriented day clinic treatment (8 am-3:30 pm, Mon–Friday, 2 months)
TAU NA 1. ISMI 2. WHOQOL-BREF
1. Significant group effect on ISMI subscale, discrimination experience and social withdrawal, which internalized stigma within control group at both baseline and follow-up were higher. 2. Significant group, time, and group by time effect in QoL of experimental group and improved in greater extent over time. 3. Significant time, group by time interaction in which significant reduction in psychopathology over time in both groups.
Starin et al., 2013
Uncontrolled clinical trial; Netherlands
Schizophrenia Schizoaffective disorder Schizophreniform disorder Psychotic disorder NOS
CBT-n (45-min session, once a week, 20 sessions)
NA NA 1. PANSS 2. A combined score of BDI-4, 10, 12, 13, 15, 19, 20, 21; BHS-2, 6 inverted, 17; DAS-DPA-6, 14, 15; ISMI
1. Time effect: the improvement on the primary outcome measure was highly significant and clinically relevant.
Uchino et al., 2012
Controlled clinical trial; Japan
Schizophrenia Schizoaffective disorder
EG: 29/29 CG: 27/27
Weekly psychoeducation program (Once a week, 6 weeks)
TAU NA 1. SDS-J 2. KIDI 3. DAI-10 4. BPIS 5. GAF
1. Group effect: significant change between 2 groups for the SDS-J, KIDI and BPIS. 2. Correlation: significant correlation found between SDS-J and KIDI, BPIS and DAI-10. 3. No significant difference for DAI-10, GAF, age, and duration of treatment.
Yanos et al., 2012a, 2012b
RCT; U.S. Schizoaffective Schizophrenia Bipolar I Bipolar II Major depression
EG: 21/15 CG: 18/NR
NECT (1-h session, once a week, 20 sessions)
TAU 3 months 1. ISMI 2. BHS 3. RSES 4. CSC 5. QLS 6. PANSS 7. SUMD
1. No significant differences between the NECT and TAU groups.
CG: Control Group, EG: Experimental Group, NA: Non Applicable; ACSS: Approaches to Copingwith Stigma Scale, ADHS: Adult Dispositional Hope Scale, BDI: Beck Depression Inventory, BHS: The Beck Hopelessness Scale, BPIS: Birchwood’s Psychosis Insight Scale, BPRS: The Brief Psychiatric Rating Scale, CAS: Cognitive Appraisal of Stigma, CAQ-SPMI: The Change Assessment Questionnaire for People with Severe and Persistent Mental Illness, CBT-n: Cognitive Behavioral Therapy for negative symptoms, CES-D: Center for Epidemiological Studies Depression scale, CGSS: The Chinese General Self-efficacy Scale, COMIS: Coming Out with Mental Illness Scale, CSC: The Coping with Symptoms Checklist, CSSMIS: The Chinese Self-stigma of Mental Illness Scale, DAI-10: Drug Attitude Inventory, DAS-DPA: Dysfunctional Attitude Scale—Defeatist Performance Attitude, GAF: Global Assessment of Functioning Scale, GPSES: Generalized Perceived Self-Efficacy Scale, ISMI: Internalized Stigma ofMental Illness Scale, KIDI: Knowledge of Illness andDrugs Inventory, LPSQ: Link Perceived Stigma Questionnaire, MANSA: TheManchester Short Assessment of Quality of Life, MES: Modified Engulfment Scale, MHRM: TheMental Health Recovery Measure, MSEI: Multidimensional Self-Esteem Inventory, MSPSS:Multidimensional Scale of Perceived Social Support, NECT: Narrative Enhancement/Cognitive Therapy, PANSS: The Positive and Negative Syndrome Scale, PGRS: Personal Growth and Recovery Scale, PTCS: The Psychosocial Treatment Compliance Scale, QLS: The Quality of Life Scale, RSES: Rosenberg Self-Esteem Scale, SDS-J: Japanese version of Social Distance Scale, SES: Self-Efficacy Scale, SUMD: The Scale for Assessing Unaware- ness of Mental Disorder, TAU: Treatment as usual, TSCS: Tennessee Self-Concept Scale. 51
Ta bl e.
2 Q ua
lit y A ss es sm
en t of
th e in cl ud
ed st ud
om iz at io n
em pl oy
iz at io n m et ho
ds In cl us
an d ex
cr it er ia
Pa ti en
t Bl in de
d O ut co
as se ss or s
bl in de
W it hd
ra w al
an d dr op
Sa m pl e
es ti m at io n
In te nt io n-
to -t re at
an al ys is
In te rv en
de sc ri be
d Q ua
lifi ca ti on
of th e tr ai ne
Co rr ig an
et al ., 20
N Y (i nc
lu si on
on ly )
Co st ai n et
al ., 20
14 N A
da r an
di gi t nu
r ta bl e
et al ., 20
er at io n of
et al ., 20
11 N A
Y (I nc
lu si on
on ly )
Ly sa ke
al ., 20
12 N A
Y N A
al ., 20
By ra nd
pu lli ng
p as si gn
t( gr ou
co nt ro lf ro m
ve lo pe
Ru ss in ov
al ., 20
Co m pu
te r- ge
ne ra te d pr og
st ra ti fie
ra ci al -e th ni c m in or it y st at us
Ro e et
al ., 20
sc h et
al ., 20
r ta bl e,
an d en
ve lo pe
s w it h nu
rs w er e
re se ar ch
as si st an
ts w ho
ha d no
e w hi ch
nu m be
r w as
ve lo pe
Si bi tz
et al ., 20
St ar in g et
al ., 20
13 N A
Y U nb
Y U ch
in o et
al ., 20
12 N o
al ., 20
ra te d ra nd
om iz at io n fo r tw
o se tt in gs
ba se d on
co m pu
te ri ze
ne ra ti ng
sy st em
ot re po
;N A :n
pl ic ab
:y es .
52 H.W.H. Tsang et al. / Schizophrenia Research 173 (2016) 45–53
more attention. In future, more innovative approaches to reducing in- ternalized stigma should be developed andmore RCTs on particular in- tervention components using standard outcome measure should be conducted so that meta-analysis could be conducted and effects of the intervention could be compared. All of the above adds to evidence-based practice in internalized stigma reduction.
Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.schres.2016.02.013.
Cohen, J., 1988. Statistical Power Analysis for the Behavioral Sciences. second ed. Law- rence Erlbaum Associates, Hillsdale, NJ.
Corrigan, P.W., 2014. The impact of mental illness stigma on seeking and participating in mental health care. Psychol. Sci. Public Interest 15 (2), 37–70. http://dx.doi.org/10. 1177/1529100614531398.
Corrigan, P.W., et al., 2015. Diminishing the self-stigma of mental illness by coming out proud. Psychiatry Res. 229 (1–2), 148–154. http://dx.doi.org/10.1016/j.psychres. 2015.07.053.
Costain, G., Esplen, M.J., Toner, B., Scherer, S.W., Meschino, W.S., Hodgkinson, K.A., Bassett, A.S., 2014. Evaluating Genetic Counseling for Individuals With Schizophre- nia in the Molecular Age. Schizophr. Bull. 40 (1), 78–87. http://dx.doi.org/10.1093/ schbul/sbs138.
Çuhadar, D., Çam, M.O., 2014. Effectiveness of psychoeducation in reducing internalized stigmatization in patients with bipolar disorder. Arch. Psychiatr. Nurs. 28 (1), 62–66. http://dx.doi.org/10.1016/j.apnu.20 13.10.008.
Fung, K.M., Tsang, H.W., Cheung, W.M., 2011. Randomized controlled trial of the self- stigma reduction program among individuals with schizophrenia. Psychiatry Res. 189 (2), 208–214. http://dx.doi.org/10.1016/j.psychres.2011.02.013.
Fung, K.M., Tsang, H.W., Corrigan, P.W., 2008. Self stigma of peoplewith schizophrenia as predictor of their adherence to psychosocial treatment. Psychiatr. Rehabil. J. 32 (2), 95–104.
Gerlinger, G., et al., 2013. Personal stigma in schizophrenia spectrum disorder: a system- atic review of prevalence rates, correlates, impact and interventions. World Psychi- atry 12 (2), 155–164. http://dx.doi.org/10.1002/wps.20040.
Hayward, P., Bright, J., 1997. Stigma and mental illness: a review and critique. J. Ment. Health 6 (4), 345–354.
Higgins, J.P.T., Green, S., 2011. Cochrane handbook for systematic reviews of interven- tions version 5.1.0. The Cochrane Collaboration (Retrieved from www.cochrane- handbook.org).
Higgins, J.P.T., Thompson, S.G., Deeks, J.J., Altman, D.G., 2003. Measuring inconsistency in meta-analyses. BMJ 327 (7414), 557–560. http://dx.doi.org/10.1136/bmj.327.7414. 557.
Lacro, J.P., Dunn, L.B., Dolder, C.R., Leckband, S.G., Jeste, D.V., 2002. Prevalence of and risk factors for medication nonadherence in patients with schizophrenia: a comprehen- sive review of recent literature. J. Clin. Psychiatry 63 (10), 892–909.
Liberati, A., Altman, D.G., Tetzlaff, J., Mulrow, C., Gøtzsche, P.C., Loannidis, J.P.A., … Moher, D., 2009. The PRISMA statement for reporting systematic reviews andmeta-analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS Med. 6 (7), 1–28. http://dx.doi.org/10.1016/j.jclinepi.2009.06.006.
Livingston, J., Boyd, J., 2010. Correlates and consequences of internalized stigma for peo- ple living with mental illness: a systematic review and meta-analysis. Soc. Sci. Med. 71 (12), 2150–2161. http://dx.doi.org/10.1016/j.socscimed.2010.09.030.
Lucksted, A., Drapalski, A., Calmes, C., Forbes, C., DeForge, B., Boyd, J., 2011. Ending self- stigma: pilot evaluation of a new intervention to reduce internalized stigma among people with mental illnesses. Psychiatr. Rehabil. J. 35 (1), 51–54. http://dx. doi.org/10.2975/35.1.2011.51.54.
Lysaker, P.H., Roe, D., Ringer, J., Gilmore, E.M., Yanos, P.T., 2012. Change in self-stigma among persons with schizophrenia enrolled in rehabilitation: associations with self-esteem and positive and emotional discomfort symptoms. Psychol. Serv. 9 (3), 240–247. http://dx.doi.org/10.1037/a0027740.
McCay, E., Beanlands, H., Zipursky, R., Roy, P., Leszcz, M., Landeen, J., … Chan, E., 2007. A randomised controlled trial of a group intervention to reduce engulfment and self- stigmatisation in first episode schizophrenia. Adv. Ment. Health 6 (3), 212–220. http://dx.doi.org/10.5172/jamh.6.3.212.
Mittal, D., Sullivan, G., Chekuri, L., Allee, El, Corrigan, P.W., 2012. Empirical studies of self- stigma reducation strategies: a critical review of the literature. Psychiatr. Serv. 63 (10), 974–981.
Mosanya, T.J., Adelufosi, A.O., Adebowale, O.T., Oqunwale, A., Adebayo, O.K., 2014. Self- stigma, quality of life and schizophrenia: An outpatient clinic survey in Nigeria. Int. J. Soc. Psychiatry 60 (4), 377–386. http://dx.doi.org/10.1177/0020764013491738.
Roe, D., Hasson-Ohayon, I., Mashiach-Eizenberg, M., Derhy, O., Lysaker, P.H., Yanos, P.T., 2014. Narrative enhancement and cognitive therapy (NECT) effectiveness: a quasi- experimental study. J. Clin. Psychol. 70 (4), 303–312. http://dx.doi.org/10.1002/ jclp.22050.
Rüsch, N., Abbruzzese, E., Hagedorn, E., Hartenhauer, D., Kaufmann, I., Curschellas, J., … Corrigan, P.W., 2014. Efficacy of coming out proud to reduce stigma’s impact among people with mental illness: pilot randomised controlled trial. Br. J. Psychiatry 204, 391–397. http://dx.doi.org/10.1192/bjp.bp.113.135772.
Rüsch, N., Corrigan, P.W., Wassel, A., Michaels, P., Larson, J.E., Olschewski, M., Wilkniss, S., Batia, K., 2009. Self-stigma, group identification, perceived legitimacy of discrimina- tion and mental health service use. Br. J. Psychiatry 195 (6), 551–552. http://dx.doi. org/10.1192/bjp.bp.109.067157.
53H.W.H. Tsang et al. / Schizophrenia Research 173 (2016) 45–53
Russinova, Z., Rogers, E.S., Gagne, C., Bloch, P., Drake, K.M., Mueser, K.T., 2014. A random- ized controlled trial of a peer-run antistigma photovoice intervention. Psychiatr. Serv. 65 (2), 242–246. http://dx.doi.org/10.1176/appi.ps.201200572.
Sibitz, I., Provaznikova, K., Lipp, M., Lakeman, R., Amering, M., 2013. The impact of recovery-oriented day clinic treatment on internalized stigma: preliminary report. Psychiatry Res. 209 (3), 326–332. http://dx.doi.org/10.1016/j.psychres.2013.02.001.
Staring, A.B.P., Ter Huurne, M.B., van, D.G., 2013. Cognitive behavioral therapy for negative symptoms (CBT-n) in psychotic disorders: a pilot study. J. Behav. Ther. Exp. Psychia- try 44 (3), 300–306. http://dx.doi.org/10.1016/j.jbtep.2013.01.004.
The Cochrane Collaboration, 2014. Reviewmanager (RevMan) [computer program]. Ver- sion 5.3 Retrieved from http://tech.cochrane.org/revman.
Tsang, H.W.H., Fung, K.M.T., Chung, R.C.K., 2010. Self-stigma and stages of change as pre- dictors of treatment adherence of individuals with schizophrenia. Psychiatry Res. 180 (1), 10–15.
Uchino, T., Maeda, M., Uchimura, N., 2012. Psychoeducation may reduce self-stigma of people with schizophrenia and schizoaffective disorder. Kurume Med. J. 59 (1–2), 25–31.
Yanos, P.T., Roe, D., Markus, K., Lysaker, P.H., 2008. Pathways between internalized stigma and outcomes related to recovery in schizophrenia spectrum disorders. Psychiatr. Serv. 59 (12), 1437–1442.
Yanos, P.T., Roe, D., West, M.L., Smith, S.M., Lysaker, P.H., 2012a. Group-based treatment for internalized stigma among persons with severe mental illness: findings from a randomized controlled trial. Psychol. Serv. 9 (3), 248–258. http://dx.doi.org/10. 1037/a0028048.
Yanos, P.T., West, M., Gonzales, L., Smith, S., Roe, D., Lysaker, P., 2012b. Change in internal- ized stigma and social functioning among persons diagnosed with severe mental ill- ness. Psychiatry Res. 200 (2–3), 1032–1034. http://dx.doi.org/10.1016/j.psychres. 2012.06.017.
- Therapeutic intervention for internalized stigma of severe mental illness: A systematic review and meta-„analysis
- 1. Introduction
- 2. Methods
- 2.1. Data sources
- 2.2. Study selection
- 2.3. Data extraction and analysis
- 2.4. Assessment of methodological quality
- 2.5. Data synthesis and analysis
- 3. Results
- 3.1. Results of literature search
- 3.2. Description of included studies
- 3.3. Results of individual studies
- 3.4. Effects of intervention approaches on internalized stigma
- 3.4.1. Comparison 1: psychoeducation element
- 3.4.2. Comparison 2: cognitive behavioral therapy element
- 3.4.3. Comparison 3: group discussion element
- 3.4.4. Comparison 4: social skills training element
- 3.4.5. Comparison 5: narrative enhancement and cognitive therapy (NECT) element
- 3.4.6. Other intervention approaches
- 3.5. Meta-analysis
- 4. Discussion
- 5. Conclusion