Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Jan 1.
Published in final edited form as: Drug Alcohol Depend. 2019 Oct 28;206:107677. doi: 10.1016/j.drugalcdep.2019.107677

Factorial validity of a substance-use stigma scale in methamphetamine-using adults in China

Shubao Chen a,b,c,d,e,f, Yuejiao Ma a,b,c,d,e, Weifu Cai g, Tania Moretta h, Xuyi Wang a,b,c,d,e,*, Tieqiao Liu a,b,c,d,e,*, Marc N Potenza f,i,j,k,l
PMCID: PMC6980741  NIHMSID: NIHMS1543142  PMID: 31732296

Abstract

Background:

Despite the large number of individuals who use substances of abuse and the stigma of such behaviors in China, scales have not been developed and validated for assessing substance-use stigma. Given its importance for targeting interventions, the aim of the present study was to validate a Chinese substance-use stigma measure including three dimensions of substance-use-disorder-related stigma (personal stigma, perceived stigma and social distance) by modifying a pre-existing scale measuring mental illness-related stigma and social distance.

Methods:

A convenience sample of eight-hundred-and-twelve individuals with substance abuse (aged 18-68 years) from different drug rehabilitation centers completed a self-reported survey including their demographics and drug use, and the stigma and social distance scales. The adapted, translated from English to Chinese, and back-translated scales were an 18-items stigma scale including personal stigma and perceived stigma subscales and a 5-items one domain social distance scale. Confirmatory factor analysis and multigroup confirmatory factor analysis tested factorial validity and measurement invariance of the scales, respectively.

Results:

For the stigma scale, by deleting one low correlation dimension (named “weak-not-sick”), personal stigma and perceived stigma showed acceptable fit indices and internal consistency with two dimensions (named “dangerous/unpredictable” and “social discrimination”), separately. The social distance scale showed good fit indices and internal consistency as a separate facet of stigma-related substance-use problems. Invariance of the model across drug-rehabilitation centers was found.

Conclusions:

The Chinese substance-use stigma scale may serve as a valuable tool for better understanding substance-use stigma among adults in China who abuse substances.

Keywords: Stigma, Substance use, Factorial validation, China

1. Introduction

Substance-use disorders (SUDs) have been described as chronic, relapsing illnesses characterized by compulsive drug use despite adverse consequences (Rinaldi et al. 1988). By the end of 2017, 2.553 million individuals in China were registered as having abused drugs, representing an increase of 1.9% over the prior year (NNCC 2017). Despite this growth, a large gap exists between the number of individuals with these disorders and the number receiving treatment. In order to reduce barriers to treatment, a major priority has been proposed to understand and decrease stigma related to SUDs (McFarling et al. 2011, NIDA 2016).

Stigma was defined by Goffman in 1963 as an “attribute that is deeply discrediting” and reduced the bearer “from a whole and usual person to a tainted, discounted one” (Goffman 1963). Since then, a stigma conceptualization has been ascribed to multiple circumstances including urinary incontinence (Sheldon and Caldwell 1994), leprosy (Opala and Boillot 1996), cancer (Fife and Wright 2000), and mental disorders (Jo C. Phelan et al. 2000, Corrigan et al. 2004, Griffiths et al. 2006, Calear et al. 2011). Health-related stigma has been described as involving the co-occurrence of labeling, stereotyping, separating, status loss, and discrimination related to a specific disease in the context of power imbalance (Link and Phelan 2001). Multiple different facets of mental illness stigma have been identified (Rüsch et al. 2005), including personal stigma, perceived stigma, internalized self-stigma, perceived dangerousness, desire for social control, and desire for social distance (Jorm and Oh 2009). These facets may also apply to substance-use stigma.

In China, stigmatization toward people with SUDs is a considerable problem (Luo et al. 2014, van Boekel et al. 2013), and it is often more severe than stigma related to other psychiatric disorders (Corrigan et al. 2005, Corrigan et al. 2006). Negative consequences associated with stigma for SUDs are notable, including delayed treatment-seeking (Semple et al. 2005, McFarling et al. 2011), reduced mental and physical health (Ahern et al. 2007), delayed recovery (Luoma et al. 2007), poorer social relationships, unemployment, and increased shame among individuals with substance abuse and their family members (Corrigan et al. 2006). Despite the potential impact of SUD-related stigma, research on this area is limited, particularly in China.

Stigma may arise from the convergence of interrelated components that include labeling (highlighting and defining human differences), stereotyping (using labels focusing on undesirable elements), separating (using labels describing individuals as belonging to “out-groups”), and discriminating (using labels to generate status loss), with a power structure allowing these processes to exist and propagate (Link and Phelan 2006). Few studies have investigated SUD-related stigma in China. The negative attitudes towards individuals with SUDs have been found to be widespread, and most Chinese people have labeled an individual with a SUD as an “addict”, rated such individuals as having undesirable characteristics, and reported a strong desire for social distance (Link and Phelan 2006). Cultural factors have been implicated in the internalization of SUD-related stigma (Mak et al. 2015). However, the lack of a reliable validated instrument in Chinese makes it difficult to examine the impact of stigma towards substance abuse in people with substance-use problems in China.

Relatively few SUD-related stigma measures with acceptable psychometric properties exist. The Perceived Stigma of Addiction Scale (PSAS) (Luoma et al. 2010) is an 8-item scale that was adapted from an initial 12-item scale measuring perceived stigma toward serious mental illness (Link et al. 1997). In the revised scale, “someone who has been treated for substance use” replaced references to mental illness, and the scale was administered to a sample of individuals in treatment for substance abuse to measure perceived stigma among them. By deleting three low-fit rating items and one low-correlation item, the final PSAS consisted of eight items with one factor. While the scale has demonstrated acceptable internal consistency (α=0.73) (Luoma et al. 2010), it assesses only perceived stigma and may not adequately reflect other dimensions of the stigma of addiction. Brown (Brown 2011) measured three components of substance-use-related stigma using the Social Distance Scale for Substance Users (SDS-SU), Dangerousness Scale for Substance Users (DS-SU) and Affect Scale for Substance Users (AS-SU), which were adapted from measures of stigma related to mental illness (Link et al. 1999, Penn et al. 1994). College students completed these three measures, and the SDS-SU and AS-SU, but not the DS-SU, demonstrated good internal consistency and limited redundancy with the mental-illness versions (Brown 2011). However, Brown’s study focused on the use of alcohol and marijuana, but not other more severe form of substance use, such as heroin and methamphetamine use. The Substance Use Stigma Mechanisms Scale (SU-SMS) is an 18-item scale which quantifies enacted, anticipated and internalized substance-use stigma mechanisms among individuals with a range of SUDs (Smith et al. 2016). The SU-SMS assesses enacted and anticipated stigma from relevant stigma sources (i.e., family members and healthcare workers); however, as this scale was tested in a moderately sized sample, its stability warrants further investigation in a larger sample.

The aim of this study was to generate psychometrically acceptable SUD-related stigma measures in Chinese, which were adapted from the Depression Stigma Scale (DSS). (Griffiths et al. 2004) and the social distance scale (Link et al. 1999), in order to assess stigma for SUDs in individuals with substance-use problems. The DSS assessed two stigma domains, personal stigma and perceived stigma. Personal stigma reflects an individual’s own stigmatizing attitudes, and perceived stigma reflects the perception that members of a stigmatized group have about the stigmatizing attitudes and actions that the society may have towards them (Link 1987). The DSS was first created with the initial aim to assess stigmatizing attitudes among people who screened positive for depression. Items were derived from major recurring broad themes extracted from websites on depression and stigma. Principal component analysis showed that the scale had two facets, personal stigma and perceived stigma. Social distance reflects the desire to avoid contact with a particular group of people (i.e., mental illness). The DSS and social distance scale were then well used in measuring stigmatizing attitudes towards mental disorders in a wide range of population (e.g. high school students, general population, professionals) (Nicola J Reavley and Jorm. 2011, Yoshioka et al. 2014, Reavley et al. 2014) across different cultures (Griffiths et al. 2006). Personal stigmatizing attitudes in individuals with mental disorders may reduce willingness for help-seeking (Schomerus et al. 2012, Griffiths et al. 2011), perceived stigma may lead to treatment discontinuation and reduced social interactions with others (Perlick et al. 2001, Sirey et al. 2001). These adverse effects may also apply to individuals with substance-use problems. Of note, social distance between how individuals with substance abuse perceive others with SUDs could be a fundamental aspect to consider within the social motives that drive behavior toward use of substances, as it has been argued that peer influence are main factors for initiation and maintenance of SUDs (Sherman et al. 2008, Kandel et al. 1992). Therefore, understanding the attitudes that individuals with substance abuse have toward SUDs and the attitudes they perceive from others, and the social distance between how individuals with substance abuse perceive others with SUDs could potentially better help target interventions that promote treatment initiation and engagement, recovery and social integration.

Hence, we sought to evaluate on a sample of Chinese individuals with substance abuse, three dimensions of SUD-related stigma (personal stigma, perceived stigma and social distance) by modifying a pre-existing measure of mental illness-related stigma and social distance.

2. Material and methods

2.1. Participants and procedure

Eight-hundred-and-twelve individuals (695 male, 117 females, age range: 18-68 years) participated. They were recruited from three voluntary drug rehabilitation centers (two located in Changsha, one in Wuhan) and two compulsory drug rehabilitation centers (one in Changsha, one in Zhangzhou). All participants were provided information about the study, and were free to decline participation before they completed the survey. Oral consent to participate was obtained from each participant before they completed the survey.

A paper-based self-reported survey was provided to participants who gave their oral consent. Participants were informed that any information we obtained may be shared only with the head of research and the researchers involved in this study, who are bound by professional secrecy and required to comply with the legislation on the privacy protection. The survey was conducted between December 2015 and March 2018. The study was approved by the ethical review board of the Second Xiangya Hospital of Central South University, the ethical review board evaluated the study specifically related to participation of incarcerated individuals, and conditions for use of incarcerated individuals in research were met.

2.2. Measures

At the beginning of the questionnaire, participants were asked to provide information regarding their demographics and drug use (e.g., age of first-time methamphetamine use, they had used methamphetamine for how long, and how many times they had relapsed). The stigma and social distance scales we used were adapted and translated into Chinese by three of the authors, and back translated into English by a separate bilingual psychiatrist to help ensure the accuracy of the translation.

2.2.1. Stigma scale

Participants’ stigma attitudes to SUDs were assessed by the stigma scale comprised of 18 items adapted from the DSS (Griffiths et al. 2004). The DSS assessed two stigma domains, personal stigma and perceived stigma, and each stigma domain was assessed using 9 items. In our adaptation, we replaced the word “depression” with the word “substance-use disorder” as appropriate. The personal stigma questions were asked as follows:

“Lisi is a person with a substance-use disorder, the drugs he mainly used were synthetic drugs, such as methamphetamine. The next few questions will ask you about your personal attitudes toward Lisi. Please select from strongly agree to strongly disagree for each question. 1. People with a problem like Lisi’s could snap out of the problem if he/she wanted. 2. A problem like Lisi’s is a sign of personal weakness. 3. A problem like Lisi’s is not a real medical illness. 4. People with a problem like Lisi’s are dangerous. 5. It’s best to avoid people with a problem like Lisi’s so that you don’t develop this problem yourself. 6. A problem like Lisi’s makes him/her unpredictable. 7. I would not tell anyone else if I had a problem like Lisi’s. 8. I would not employ someone if I knew he/she had the problem like Lisi’s. 9. I would not vote for a politician if I knew he/she had the problem like Lisi’s.”

The questions for perceived stigma involved the same items, but instead asking the participants’ personal attitude, they were asked “what you think most other people’s beliefs”. Ratings were made on a 5-point Likert scale from “strongly agree” to “strongly disagree”. In Yoshioka’s model (Yoshioka et al. 2014), the authors found “weak not sick” (consisting of items 1~3) and “dangerous/unpredictable” (consisting of items 4~7) factors when assessing for personal and perceived stigma using an adapted version for young people (Jorm and Wright 2008). In our study, we used the complete scale with two more items on employing and voting for each stigma domain, and preliminary analyses using our sample suggested a third factor for these two items; we named this factor as “social discrimination” (consisting of item 8~9). Taken together, in the present study, we decided to test for the three-factor structure for both personal and perceived stigma. Scores on each stigma domain ranged from 0 to 36, with higher scores indicating greater stigma.

2.2.2. Social distance scale

Participants’ self-reported willingness to have contact with SUDs were assessed by a social distance scale adopted from the scale developed by Link et al. (Link et al. 1999). The items were rated according to how willing participants would be to engage in five social contexts with people with SUDs including: 1) move next door, 2) spend an evening socializing, 3) make friends, 4) work closely, and 5) marry into the family. Responses were made on a 4-point Likert scale from “definitely willing to” to “definitely unwilling to”. Scores for each item were summed and divided by 5, with total scores ranging from 1 (low social distance) to 4 (high social distance).

2.3. Statistical analysis

First, a confirmatory factor analysis (CFA) using the Lavaan package (Rosseel 2012) of R software (R Development Core Team 2018) was conducted. The fit of the general model was assessed using the following indices: Chi square (χ2), Comparative Fit Index (CFI; values between 0.95~0.97 were considered as indicating acceptable fit, greater than 0.97 as good fit), Tucker Lewis Index (TLI; values between 0.95~0.97 were considered as indicating acceptable fit, greater than 0.97 as good fit), and Root Mean Square Error of Approximation (RMSEA; values between 0.05~0.08 were considered as indicating acceptable fit, lower than 0.05 as good fit) (Schermelleh-Engel et al. 2003). Cronbach’s α was used to assess internal consistencies of the scales and its dimensions (values between 0.70~0.80 were considered as acceptable, between 0.80~0.90 were considered as good, and greater than 0.90 as excellent) (Streiner 2003). When finishing the first CFA, we found the Cronbach’s α was unacceptable. We then checked the correlation of each item with other items and the total score, removed the items with low correlations with other items and the total score (R<0.30) (Tavakol and Dennick 2011) and generated a new model. Another CFA was conducted with the new model.

Second, the new model was tested independently for both voluntary and compulsory drug rehabilitation centers to assess configural invariance. A multigroup CFA was conducted to assess measurement invariance of the scale across different drug rehabilitation centers. Next, a hierarchical approach was adopted to constrain model parameters and compare changes in model fit. Configural, metric and scalar models were estimated. Measurement invariance was evaluated by both the change in model fit indices (i.e., ΔCFI larger than 0.01 and ΔRMSEA lager than 0.015 were considered as indicative of non-invariance; (Cheung and Rensvold 2002)) and multigroup model fit indices.

3. Results

Of the 812 participants, 794 (97.8%) completed the stigma and social distance scales, 784 (98.8%) out of the 794 participants used methamphetamine and were included in the final analyses. Descriptive statistics of socio-demographic, drug-use measures, stigma and social distance scores, and the differences between the voluntary versus compulsory drug rehabilitation centers are shown in Table 1. Individual’s age at first-time methamphetamine use ranged from 11-50 years; the number of relapses ranged from 0-10 times; and period of methamphetamine use ranged from 1-227 months.

Table 1.

Descriptive statistics of socio-demographic, drug-use measures, stigma and social distance.

Socio-demographic characteristic All (N=784) Voluntary
(N=603)		 Compulsory
(N=181)		 F/χ2 p
Age, years (SD) 31.59 (7.16) 31.47(6.88) 31.98(8.03) 0.78 0.436
Education, years (SD) 10.89 (3.25) 11.38(3.29) 9.28(2.51) −9.13 <0.001
Gender 173.53 <0.001
 Male, N (%) 676 (86.22) 574(95.19) 102(56.35)
 Female, N (%) 108 (13.78) 29(4.81) 79(43.65)
Marriage 32.89 <0.001
 Currently married, N (%) 356 (45.41) 308(51.08) 48(26.52)
 Not married, N (%) 428 (54.59) 295(48.92) 133(73.48)
Drug use measure
Age at first-time methamphetamine use, years (SD) 24.19 (7.09) 24.00(6.94) 24.82(7.57) 1.30 0.194
 Number of relapses, mean (SD) 1.36 (1.62) 1.29(1.60) 1.61(1.68) 2.26 0.025
 Period of methamphetamine use, months (SD) 47.29 (36.31) 49.12(36.68) 41.19(34.45) −2.67 0.008
Stigma
 Personal stigma, mean (SD) 22.92 (3.75) 22.81(3.77) 23.28(3.67) 1.51 0.131
 Perceived stigma, mean (SD) 24.35 (3.56) 24.46(3.56) 23.96(3.55) −1.69 0.092
 Social distance, mean (SD) 3.04 (0.77) 3.05(0.81) 3.03(0.65) −0.26 0.794
Open in a new tab

Results of the CFA for the stigma general model (model 1) showed a good fit index but poor internal consistency with respect to the observed data (personal stigma: χ2(24)=54.878, p<0.001, CFI=0.971, TLI=0.957, RMSEA=0.041 90%CI=0.026 0.055, Cronbach’s α=0.36; and perceived stigma: χ2(24)=100.875, p<0.001, CFI=0.966, TLI=0.949, RMSEA=0.064 90%CI=0.051 0.077, Cronbach’s α=0.50) (Table 2). We then computed the correlation of each item with other items and the total score for each stigma (Table 3 for personal stigma, Table 4 for perceived stigma), found that the first three items (consisting of the “weak-not-sick” factor) had low correlations with other items and the total score (R<0.30). The first three items were removed and a new model generated with two factors (model 2). Results of the CFA for stigma model 2 showed a good fit index and internal consistency (personal stigma: χ2(8)=15.612, p=0.048, CFI=0.992, TLI=0.985, RMSEA=0.035 90%CI=0.003 0.060, Cronbach’s α=0.71; and perceived stigma: χ2(8)=20.078, p=0.010, CFI=0.994, TLI=0.989, RMSEA=0.044 90%CI=0.020 0.068, Cronbach’s α=0.85) (Table 2). Standardized loadings in model 2 ranged between 0.23 and 0.82 for personal stigma and 0.43 and 0.88 for perceived stigma (Table 5).

Table 2.

Fit indices for the CFA model

Scales χ2 (p) df CFI TLI RMSEA (90%CI) Cronbach’s α
Model 1
 Personal stigma 54.878 (<0.001) 24 0.971 0.957 0.041 (0.026,0.055) 0.36
 Perceived stigma 100.875 (<0.001) 24 0.966 0.949 0.064 (0.051,0.077) 0.50
Model 2
 Personal stigma 15.612 (0.048) 8 0.992 0.985 0.035 (0.003,0.060) 0.71
 Perceived stigma 20.078 (0.010) 8 0.994 0.989 0.044 (0.020,0.068) 0.85
Social distance 40.233 (<0.001) 5 0.985 0.971 0.095 (0.069,0.123) 0.89
Open in a new tab

Table 3.

Correlation of each item with other items and the total score of personal stigma scale

Variable 1 2 3 4 5 6 7 8 9 10
1.People could snap out of this kind of problem -
2.This kind of problem is a sign of personal
weakness		 0.04 -
3.This kind of problem is not a real medical
illness		 0.11 0.18 -
4.People with this kind of problem are dangerous −0.15 −0.19 −0.19 -
5.It’s best to avoid people with this kind of problem −0.12 −0.11 −0.13 0.43 -
6.People with this kind of problem are unpredictable −0.14 −0.16 −0.14 0.36 0.44 -
7.I would not tell anyone if I had this kind of problem −0.11 −0.04 −0.05 0.06 0.16 0.15 -
8.I would not employ anyone with this kind of problem −0.13 −0.17 −0.04 0.30 0.38 0.32 0.20 -
9.I would not vote anyone with this kind of problem −0.13 −0.13 −0.06 0.27 0.33 0.30 0.16 0.61 -
10.Total personal stigma score 0.15 0.13 0.21 0.50 0.61 0.53 0.40 0.63 0.59
Mean 2.31 1.82 2.08 2.68 2.83 2.78 2.68 2.87 2.96 22.92
SD 1.19 1.04 1.08 1.09 0.99 0.91 1.03 0.99 0.91 3.75
Open in a new tab

Table 4.

Correlation of each item with other items and the total score of perceived stigma scale

Variable 1 2 3 4 5 6 7 8 9 10
1.People could snap out of this kind of problem -
2.This kind of problem is a sign of personal
weakness		 0.01 -
3.This kind of problem is not a real medical
illness		 0.18 0.13 -
4.People with this kind of problem are dangerous −0.22 −0.20 −0.10 -
5.It’s best to avoid people with this kind of problem −0.26 −0.10 −0.06 0.62 -
6.People with this kind of problem are unpredictable −0.30 −0.15 −0.10 0.62 0.63 -
7.I would not tell anyone if I had this kind of problem −0.23 0.08 −0.05 0.30 0.32 0.38 -
8.I would not employ anyone with this kind of problem −0.24 −0.09 −0.07 0.48 0.57 0.52 0.30 -
9.I would not vote anyone with this kind of problem −0.26 −0.12 −0.07 0.46 0.56 0.51 0.28 0.75 -
10.Total perceived stigma score 0.03 0.20 0.28 0.62 0.69 0.65 0.53 0.69 0.66 -
Mean 1.49 1.67 2.02 3.28 3.29 3.19 3.05 3.14 3.21 24.35
SD 1.10 1.01 1.01 0.77 0.79 0.79 0.86 0.81 0.77 3.56
Open in a new tab

Table 5.

Standardized factor loadings for the stigma and social distance scale

Items Weak-not-sick Dangerous/
unpredictable		 Social
discrimination		 Social
distance
Model 1
 Personal stigma
 1. People could snap out of this kind of problem 0.28 - - -
 2. This kind of problem is a sign of personal weakness 0.37 - - -
 3. This kind of problem is not a real medical illness 0.37 - - -
 4. People with this kind of problem are dangerous - 0.60 - -
 5. It’s best to avoid people with this kind of problem - 0.70 - -
 6. People with this kind of problem are unpredictable - 0.63 - -
 7. I would not tell anyone if I had this kind of problem - 0.23 - -
 8. I would not employ anyone with this kind of problem - - 0.82 -
 9. I would not vote anyone with this kind of problem - - 0.74 -
 Perceived stigma
 1. People could snap out of this kind of problem 0.55 - - -
 2. This kind of problem is a sign of personal weakness 0.18 - - -
 3. This kind of problem is not a real medical illness 0.26 - - -
 4. People with this kind of problem are dangerous - 0.76 - -
 5. It’s best to avoid people with this kind of problem - 0.81 - -
 6. People with this kind of problem are unpredictable - 0.80 - -
 7. I would not tell anyone if I had this kind of problem - 0.43 - -
 8. I would not employ anyone with this kind of problem - - 0.88 -
 9. I would not vote anyone with this kind of problem - - 0.86 -
Model 2
 Personal stigma
 4. People with this kind of problem are dangerous - 0.58 - -
 5. It’s best to avoid people with this kind of problem - 0.73 - -
 6. People with this kind of problem are unpredictable - 0.62 - -
 7. I would not tell anyone if I had this kind of problem - 0.23 - -
 8. I would not employ anyone with this kind of problem - - 0.82 -
 9. I would not vote anyone with this kind of problem - - 0.74 -
 Perceived stigma
 4. People with this kind of problem are dangerous - 0.76 - -
 5. It’s best to avoid people with this kind of problem - 0.81 - -
 6. People with this kind of problem are unpredictable - 0.79 - -
 7. I would not tell anyone if I had this kind of problem - 0.43 - -
 8. I would not employ anyone with this kind of problem - - 0.88 -
 9. I would not vote anyone with this kind of problem - - 0.86 -
Social distance
Move next door - - - 0.80
Spend an evening socializing - - - 0.88
Make friends - - - 0.89
Work closely - - - 0.82
Marry into the family - - - 0.56
Open in a new tab

Results of the correlation of each item with other items and the total score for social distance were showed in Table 6. Results of the CFA for social distance showed a good fit index and internal consistency (χ2(5)=40.233, p<0.001, CFI=0.985, TLI=0.971, RMSEA=0.095 90%CI=0.069 0.123, Cronbach’s α=0.89) (Table 2). Standardized loadings ranged between 0.56 and 0.89 (Table 5).

Table 6.

Correlation of each item with other items and the total score of social distance scale

Variable 1 2 3 4 5 6
1.Move next door -
2.Spend an evening socializing 0.72 -
3.Make friends 0.69 0.80 -
4.Work closely 0.66 0.70 0.73 -
5.Marry into the family 0.44 0.46 0.50 0.53 -
6.Total social distance score 0.85 0.89 0.90 0.87 0.67 -
Mean 2.92 2.85 2.92 3.00 3.53 3.04
SD 0.95 0.97 0.97 0.94 0.77 0.77
Open in a new tab

The new model (model 2) was estimated in voluntary and compulsory drug rehabilitation centers separately before testing the measurement invariance. Results (Table 7) showed that the model fit was adequate to good for both voluntary (personal stigma: χ2(8)=9.466, p=0.304; CFI=0.998; TLI=0.996; RMSEA=0.017 90%CI=0.000 0.053; perceived stigma: χ2(8)=23.231, p=0.003; CFI=0.991; TLI=0.984; RMSEA=0.056 90%CI=0.030 0.083; and social distance: χ2(5)=31.023, p<0.001; CFI=0.987; TLI=0.973; RMSEA=0.093 90%CI=0.063 0.125) and compulsory (personal stigma: χ2(8)=17.737, p=0.023, CFI=0.950, TLI=0.907, RMSEA=0.082 90%CI=0.029 0.134; perceived stigma: χ2(8)=22.865, p=0.004, CFI=0.960, TLI=0.924, RMSEA=0.101 90%CI=0.054 0.151; and social distance: χ2(5)=15.650, p=0.008, CFI=0.977, TLI=0.953, RMSEA=0.108 90%CI=0.050 0.171) drug rehabilitation centers. The fit indices of the unconstrained multigroup model indicated the configural invariance of the factor structure was similar across drug rehabilitation centers (personal stigma: χ2(16)=27.203, p=0.039, CFI=0.988, RMSEA=0.042 90%CI=0.009 0.069; perceived stigma: χ2(16)=46.096, p<0.001, CFI=0.986, RMSEA=0.069 90%CI=0.046 0.093; and social distance: χ2(10)=46.673, p<0.001, CFI=0.985, RMSEA=0.097 90%CI=0.070 0.126). Metric invariance was tested subsequently by constraining all item loadings to be equal across drug rehabilitation centers, and no significant reduction in model fit was found (personal stigma: ΔCFI=−0.001, ΔRMSEA=0.005; perceived stigma: ΔCFI=0.001, ΔRMSEA=0.006; social distance: ΔCFI=0.000, ΔRMSEA=0.017). Finally, scalar invariance was tested by constraining all item intercepts to be equal across drug rehabilitation centers (personal stigma: ΔCFI=0.001, ΔRMSEA=0.003; perceived stigma: ΔCFI=0.004, ΔRMSEA=−0.003; social distance: ΔCFI=0.002, ΔRMSEA=0.004).

Table 7.

Fit indices for measurement invariance tests on the stigma and social distance scale

Model N χ2 (p) df CFI ΔCFI RMSEA ΔRMSEA
Personal stigma
Drug rehabilitation center
 Voluntary 603 9.466(0.304) 8 0.998 - 0.017 -
 Compulsory 181 17.737(0.023) 8 0.950 - 0.082 -
Configural invariance 784 27.203(0.039) 16 0.988 - 0.042 -
Metric invariance 784 30.666(0.060) 20 0.989 −0.001 0.037 0.005
Scalar invariance 784 35.070(0.067) 24 0.988 0.001 0.034 0.003
Perceived stigma
Drug rehabilitation center
 Voluntary 603 23.231(0.003) 8 0.991 - 0.056 -
 Compulsory 181 22.865(0.004) 8 0.960 - 0.101 -
Configural invariance 784 46.096(<0.001) 16 0.986 - 0.069 -
Metric invariance 784 51.301(<0.001) 20 0.985 0.001 0.063 0.006
Scalar invariance 784 65.337(<0.001) 24 0.981 0.004 0.066 −0.003
Social distance
Drug rehabilitation center
 Voluntary 603 31.023(<0.001) 5 0.987 - 0.093 -
 Compulsory 181 15.650(<0.008) 5 0.977 - 0.108 -
Configural invariance 784 46.673(<0.001) 10 0.985 - 0.097 -
Metric invariance 784 48.733(<0.001) 14 0.985 0.000 0.080 0.017
Scalar invariance 784 59.268(<0.001) 18 0.983 0.002 0.076 0.004
Open in a new tab

4. Discussion

The aim of the present study was to evaluate the psychometric properties of a new multi-dimensional stigma measure in the context of SUDs on individuals with substance-use problems. Using the pre-existing mental illness-related stigma and social distance measures, we established a measure of stigma in the context of SUDs with different stigma sources, i.e. personal stigma, perceived stigma, and social distance, presenting acceptable psychometric properties. In Yoshioka’s model, the authors found “weak not sick” and “dangerous/unpredictable” for both personal and perceived stigma toward mental disorders, using an adapted version for young people (Yoshioka et al. 2014). Different from Yoshioka’s model, we found that in individuals with substance abuse, in the personal and perceived stigma scale, the items relating to the “weak-not-sick” dimension had low correlations with other items and the total score. After removing these items, CFA revealed good fit index and internal consistency with two dimensions (“dangerous/unpredictable” and “social discrimination”) for each stigma scale. In addition, we found that the social distance scale had a good fit index and internal consistency as a separate construct of stigma for SUDs, which replicated and extended previous findings in the context of mental disorders (Jorm and Wright 2008). Lastly, results of the multigroup analysis suggested the invariance of the model across drug rehabilitation centers.

The reason underlying the lack of identification of the “weak-not-sick” factor warrants consideration. In our study, all participants had substance-use problems and were recruited from treatment centers, and some had experienced prior relapses. It is thus likely that they could acknowledge the difficulties relating to ceasing to use drugs and the importance of seeking medical help. The current sample reported predominantly problems with methamphetamine use. Unlike the case for many other drugs (e.g., opioids), there is a tendency for individuals using methamphetamine to experience drug-related psychosis (Farrell et al. 2002), which may more readily be identified by individuals as a “sickness”. In the past two decades, the brain disease model of addiction has been accepted by many, including in China (Volkow et al. 2016). These considerations may in part explain why the “weak-not-sick” dimension did not fit well in the present study; however, direct investigation of these possibilities is warranted.

While compulsory drug rehabilitation centers are common throughout East and Southeast Asia, their effectiveness is controversial (Werb et al. 2016). Some data suggest that relapse rates of individuals released from compulsory drug rehabilitation centers are high (Wegman et al. 2017), and stigma and discrimination may represent potential precipitants of relapse (Yang et al. 2015). Results of the multigroup analysis in the present study support the potential utility of the SUD-related stigma measure in relapse prevention programs in order to better understand which facets of SUD-related stigma may relate to relapse and how to reduce SUD-related stigma in order to help individuals maintaining abstinence, thus improving health outcomes and well-being.

Despite the psychometric strengths of this new SUD-related stigma measure, several limitations of the current study should be considered. We only included SUD-related personal stigma, perceived stigma and social distance in this measure, future work could consider assessing additional facets of SUD-related stigma, such as self-esteem, internalized self-stigma and anticipated stigma. We mainly investigated individual’s stigma attitude towards synthetic drug use. Attitude towards alcohol drinking, cigarette use, heroin use, and behavioral addictions could differ, it is necessary to investigate this measure relative to other addictive behaviors. While the scales operated well with respect to psychometric properties, some aspects of stigma may not have been optimally assessed. For example, “perceived stigma” measures were focused on how “most other people” viewed SUDs, they did not assess “felt stigma” per se. While our sample size was considerable and well suited for CFA (Wolf et al. 2013), the limited number of female participants restricted us to test the measurement properties across groups. Future studies should investigate possible gender-related differences. Our sample consisted predominantly of individuals with problems of methamphetamine use, reflective of current drug-use trends in China (NNCC 2017). The extent to which the findings relate to groups with other drug-use problems or addictive behaviors warrants additional study. We studied individuals in treatment for substance-use problems. The extent to which the findings may extend to individuals with substance-use problems who are not in treatment should be examined in future studies. Thus, future work should consider validating this measure in other populations (including in community samples to determine relative sensitivity in different populations) and investigating the attitude towards other drug-use stigma, as well as examining other facets of stigma. Lastly, to our knowledge, there is no available Chinese instrument that measures SUD-related personal and perceived stigma and social distance. As such, we could not determine the extent to which different instruments measure the same construct in order to assess more fully construct validity.

5. Conclusions

The current work found this Chinese version of SUD-related stigma measure to be a reliable and valuable tool for assessing the stigmatizing attitudes towards SUDs, particularly towards synthetic drugs, in individuals with substance-use problems. Notwithstanding the above limitations, this measure is positioned as a useful tool in assessing the clinical relevance of stigma as related to SUDs and developing promising interventions focused on reducing stigma and its impact. Such efforts are critical for reducing treatment barriers and optimizing relapse prevention programs in order to improve social and health outcomes.

Highlights.

  • Personal and perceived stigma may influence people with stimulant-use disorders.

  • The Chinese substance-use stigma scale had acceptable fit indices and internal consistency.

  • The scale had dimensions of personal and perceived stigma.

  • A social distance scale had a good fit index and internal consistency as a separate domain.

  • Multi-group analysis suggested scale invariance across drug-rehabilitation centers.

Acknowledgments

Role of Funding Source

This work was supported by National Key R & D Program of China (2017YFC1310400) and Natural Science Foundation of China (81371465, 81671324, 81571307), Fundamental Research Founds for the Central Universities of Central South University (2016zzts148), NIH grants R01 DA039136 and P50 DA09241.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Conflict of interest

The authors declare no conflicts of interest with the content of the manuscript.

References

  1. Ahern J, Stuber J and Galea S 2007. Stigma, discrimination and the health of illicit drug users. Drug Alcohol Depend., 88, 188–196. [DOI] [PubMed] [Google Scholar]
  2. Brown SA 2011. Standardized measures for substance use stigma. Drug Alcohol Depend., 116, 137–141. [DOI] [PubMed] [Google Scholar]
  3. Calear AL, Griffiths KM and Christensen H 2011. Personal and perceived depression stigma in Australian adolescents: magnitude and predictors. J. Affect. Disord, 129, 104–108. [DOI] [PubMed] [Google Scholar]
  4. Cheung GW and Rensvold RB 2002. Evaluating goodness-of-fit indexes for testing measurement invariance. Struct. Equ. Modeling, 9, 233–255. [Google Scholar]
  5. Corrigan PW, Lurie BD, Goldman HH, Slopen N, Medasani K and Phelan S 2005. How adolescents perceive the stigma of mental illness and alcohol abuse. Psychiatr Serv., 56, 544–550. [DOI] [PubMed] [Google Scholar]
  6. Corrigan PW, Markowitz FE and Watson AC 2004. Structural levels of mental illness stigma and discrimination. Schizophr. Bull, 30, 481–491. [DOI] [PubMed] [Google Scholar]
  7. Corrigan PW, Watson AC and Miller FE 2006. Blame, shame, and contamination: The impact of mental illness and drug dependence stigma on family members. J. Fam. Psychol, 20, 239. [DOI] [PubMed] [Google Scholar]
  8. Farrell M, Marsden J, Ali R and Ling W 2002. Methamphetamine: drug use and psychoses becomes a major public health issue in the Asia Pacific region. Addiction, 97, 771–772. [DOI] [PubMed] [Google Scholar]
  9. Fife BL and Wright ER 2000. The demensionality of stigma a comparison of its impact on the self of persons with HIV/AIDS and cancer. J. Health Soc. Behav, 41, 50–67. [PubMed] [Google Scholar]
  10. Goffman E, 1963. Stigma: notes on the management of spoiled identity.: Prentice Hall, Englewood Cliffs, NJ. [Google Scholar]
  11. Griffiths KM, Christensen H, Jorm AF, Evans K and Groves C 2004. Effect of web-based depression literacy and cognitive–behavioural therapy interventions on stigmatising attitudes to depression : Randomised controlled trial. Br. J. Psychiatry, 185, 342–349. [DOI] [PubMed] [Google Scholar]
  12. Griffiths KM, Crisp DA, Jorm AF and Christensen H 2011. Does stigma predict a belief in dealing with depression alone? J. Affect. Disord, 132, 413–417. [DOI] [PubMed] [Google Scholar]
  13. Griffiths KM, Nakane Y, Christensen H, Yoshioka K, Jorm AF and Nakane H 2006. Stigma in response to mental disorders: a comparison of Australia and Japan. BMC Psychiatry, 6, 21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Phelan Jo C., Link Bruce G., Stueve Ann and Pescosolido BA 2000. Public conceptions of mental illness in 1950 and 1996 What is mental illness and is it to be feared. J. Health Soc. Behav, 41, 188–207. [Google Scholar]
  15. Jorm AF and Oh E 2009. Desire for social distance from people with mental disorders. Aust. N. Z. J. Psychiatry, 43, 183–200. [DOI] [PubMed] [Google Scholar]
  16. Jorm AF and Wright A 2008. Influences on young people's stigmatising attitudes towards peers with mental disorders: national survey of young Australians and their parents. Br. J. Psychiatry, 192, 144–149. [DOI] [PubMed] [Google Scholar]
  17. Kandel DB, Yamaguchi K and Chen K 1992. Stages of progression in drug involvement from adolescence to adulthood: further evidence for the gateway theory. J. Stud. Alcohol Drugs, 53, 447–457. [DOI] [PubMed] [Google Scholar]
  18. Link BG 1987. Understanding labeling effects in the area of mental disorders: an assessment of the effects of expectations of rejection. Am. Sociol. Rev, 52, 96–112. [Google Scholar]
  19. Link BG and Phelan JC 2001. Conceptualizing stigma. Annu. Rev. Sociol, 27, 363–385. [Google Scholar]
  20. Link BG and Phelan JC 2006. Stigma and its public health implications. Lancet, 367, 528–529. [DOI] [PubMed] [Google Scholar]
  21. Link BG, Phelan JC, Bresnahan M, Stueve A and Pescosolido BA 1999. Public conceptions of mental illness: labels, causes, dangerousness, and social distance. Am. J. Public Health, 89, 1328–1333. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Link BG, Struening EL, Rahav M, Phelan JC and Nuttbrock L 1997. On stigma and its consequences: evidence from a longitudinal study of men with dual diagnoses of mental illness and substance abuse. J. Health Soc. Behav, 38, 177–190. [PubMed] [Google Scholar]
  23. Luo T, Wang J, Li Y, Wang X, Tan L, Deng Q, Thakoor JPD and Hao W 2014. Stigmatization of people with drug dependence in China: a community-based study in Hunan province. Drug Alcohol Depend., 134, 285–289. [DOI] [PubMed] [Google Scholar]
  24. Luoma JB, O'Hair AK, Kohlenberg BS, Hayes SC and Fletcher L 2010. The development and psychometric properties of a new measure of perceived stigma toward substance users. Subst. Use Misuse, 45, 47–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Luoma JB, Twohig MP, Waltz T, Hayes SC, Roget N, Padilla M and Fisher G 2007. An investigation of stigma in individuals receiving treatment for substance abuse. Addict. Behav, 32, 1331–1346. [DOI] [PubMed] [Google Scholar]
  26. Mak WW, Ho CY, Wong VU, Law RW and Chan RC 2015. Cultural model of self-stigma among Chinese with substance use problems. Drug Alcohol Depend., 155, 83–89. [DOI] [PubMed] [Google Scholar]
  27. McFarling L, D'Angelo M, Drain M, Gibbs DA and Rae Olmsted KL 2011. Stigma as a barrier to substance abuse and mental health treatment. Mil. Psychol, 23, 1–5. [Google Scholar]
  28. Reavley Nicola J and Jorm AF 2011. National survey of mental health literacy and stigma. Department of Health and Ageing, Canberra. [DOI] [PubMed] [Google Scholar]
  29. NIDA, 2016. National institute on drug abuse 2016-2020 strategic plan.
  30. NNCC, 2017. Annual report on drug control in China. Beijing: : Ministrey of Public Security. [Google Scholar]
  31. Opala J and Boillot F 1996. Leprosy among the Limba: illness and healing in the context of world view. Soc. Sci. Med, 42, 3–19. [DOI] [PubMed] [Google Scholar]
  32. Penn DL, Guynan K, Daily T, Spaulding WD, Garbin CP and Sullivan M 1994. Dispelling the stigma of schizophrenia: what sort of information is best? Schizophr. Bull, 20, 567–578. [DOI] [PubMed] [Google Scholar]
  33. Perlick DA, Rosenheck RA, Clarkin JF, Sirey JA, Salahi J, Struening EL and Link BG 2001. Stigma as a barrier to recovery: adverse effects of perceived stigma on social adaptation of persons diagnosed with bipolar affective disorder. Psychiatr. Serv, 52, 1627–1632. [DOI] [PubMed] [Google Scholar]
  34. R Development Core Team, 2018. R: A language and environment for statistical computing [online]. Available from: https://www.R-project.org/ [Accessed August 1 2018].
  35. Reavley NJ, Mackinnon AJ, Morgan AJ and Jorm AF 2014. Stigmatising attitudes towards people with mental disorders: A comparison of Australian health professionals with the general community. Aust. N. Z. J. Psychiatry, 48, 433–441. [DOI] [PubMed] [Google Scholar]
  36. Rinaldi RC, Steindler EM, Wilford BB and Goodwin D 1988. Clarification and standardization of substance abuse terminology. JAMA, 259, 555–557. [PubMed] [Google Scholar]
  37. Rosseel Y 2012. lavaan: An R package for structural equation modeling. J. Stat. Softw, 48, 1–36. [Google Scholar]
  38. Rüsch N, Angermeyer MC and Corrigan PW 2005. Mental illness stigma: concepts, consequences, and initiatives to reduce stigma. Eur. Psychiatry, 20, 529–539. [DOI] [PubMed] [Google Scholar]
  39. Schermelleh-Engel K, Moosbrugger H and Müller H 2003. Evaluating the fit of structural equation models: Tests of significance and descriptive goodness-of-fit measures. Meth. Psychol. Res. Online, 8, 33–74. [Google Scholar]
  40. Schomerus G, Auer C, Rhode D, Luppa M, Freyberger HJ and Schmidt S 2012. Personal stigma, problem appraisal and perceived need for professional help in currently untreated depressed persons. J. Affect. Disord, 139, 94–97. [DOI] [PubMed] [Google Scholar]
  41. Semple SJ, Grant I and Patterson TL 2005. Utilization of drug treatment programs by methamphetamine users: the role of social stigma. Am. J. Addict, 14, 367–380. [DOI] [PubMed] [Google Scholar]
  42. Sheldon K and Caldwell L 1994. Urinary incontinence in women: implications for therapeutic recreation. Ther. Recreation J, 28, 203–212. [Google Scholar]
  43. Sherman SG, German D, Sirirojn B, Thompson N, Aramrattana A and Celentano DD 2008. Initiation of methamphetamine use among young Thai drug users: a qualitative study. J. Adolesc. Health, 42, 36–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Sirey JA, Bruce ML, Alexopoulos GS, Perlick DA, Raue P, Friedman SJ and Meyers BS 2001. Perceived stigma as a predictor of treatment discontinuation in young and older outpatients with depression. Am. J. Psychiatry, 158, 479–481. [DOI] [PubMed] [Google Scholar]
  45. Smith LR, Earnshaw VA, Copenhaver MM and Cunningham CO 2016. Substance use stigma: Reliability and validity of a theory-based scale for substance-using populations. Drug Alcohol Depend., 162, 34–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Streiner DL 2003. Starting at the beginning: an introduction to coefficient alpha and internal consistency. J. Pers. Assess, 80, 99–103. [DOI] [PubMed] [Google Scholar]
  47. Tavakol M and Dennick R 2011. Making sense of Cronbach's alpha. Int. J. Med. Educ, 2, 53. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. van Boekel LC, Brouwers EP, van Weeghel J and Garretsen HF 2013. Stigma among health professionals towards patients with substance use disorders and its consequences for healthcare delivery: systematic review. Drug Alcohol Depend., 131, 23–35. [DOI] [PubMed] [Google Scholar]
  49. Volkow ND, Koob GF and McLellan AT 2016. Neurobiologic advances from the brain disease model of addiction. N. Engl. J. Med, 374, 363–371. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Wegman MP, Altice FL, Kaur S, Rajandaran V, Osornprasop S, Wilson D, Wilson DP and Kamarulzaman A 2017. Relapse to opioid use in opioid-dependent individuals released from compulsory drug detention centres compared with those from voluntary methadone treatment centres in Malaysia: a two-arm, prospective observational study. Lancet Glob. Health, 5, e198–e207. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Werb D, Kamarulzaman A, Meacham MC, Rafful C, Fischer B, Strathdee SA and Wood E 2016. The effectiveness of compulsory drug treatment: A systematic review. Int. J. Drug Policy, 28, 1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Wolf EJ, Harrington KM, Clark SL and Miller MW 2013. Sample size requirements for structural equation models: An evaluation of power, bias, and solution propriety. Educ. Psychol. Meas, 73, 913–934. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Yang M, Mamy J, Gao P and Xiao S 2015. From abstinence to relapse: A preliminary qualitative study of drug users in a compulsory drug rehabilitation center in Changsha, China. PLoS One, 10, e0130711. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Yoshioka K, Reavley NJ, MacKinnon AJ and Jorm AF 2014. Stigmatising attitudes towards people with mental disorders: results from a survey of Japanese high school students. Psychiatry Res., 215, 229–236. [DOI] [PubMed] [Google Scholar]

RESOURCES