Substance Use and Psychosocial Status among People Living with HIV/AIDS Who Encountered HIV Stigma in China: Stratified Analyses by Socio-Economic Status

This study examined whether the impact of HIV stigma on psychosocial status and substance use among people living with HIV/AIDS (PLWHA) differed by their socio-economic status (SES) in a Chinese setting. A total of 2,987 PLWHA were recruited from 12 sites with the highest number of cumulative HIV incidence in Guangxi, China. Participants were asked to provide information regarding their psychosocial status (e.g., depression, anxiety), history of substance use (e.g., tobacco, alcohol and drug) and SES (e.g., education, monthly income, residence type, and job category). By employing stratified multivariate regression analyses, we assessed stratum-specific impacts of HIV stigma on PLWHA’s psychosocial status and behaviors of substance use based upon participants’ SES. The impact of HIV stigma differed significantly on psychosocial status across SES gradients. Psychosocial status among people with higher education was more sensitive to HIV stigma compared with PLWHA who were less educated. The odds of substance use behaviors were higher among people with better monthly income than their low-income peers. Our study is the first paper to document the paucity of SES stratified analyses between HIV stigma and psychosocial status and substance use among PLWHA in China. We call for tailored intervention programs to target PLWHA with different backgrounds and characteristics in order to help them to better manage their seropositivity.


Introduction
By the year of 2016, it is estimated that a total of 37 million people living with HIV/AIDS (PLWHA) worldwide [1,2]. With the advancement and availability of HARRT, HIV/AIDS has transited from a lethal to a manageable chronic disease [1,2]. However, the prognosis differed significantly based upon PLWHA's socio-economic status (SES) [3,4]. SES is a multi-dimensional and complex measure that has been defined variously in history, from a measurable to an abstract construct that reflects one's access to collectively desired resources [5]. In the health related studies, SES is usually measured by educational attainment, occupation, and income levels [6].
Existing studies have indicated that low SES may be associated with higher risk of substance use [7], and the association with mental health was inconclusive [8,9]. On the other hand, studies have indicated the PLWHA with higher SES would have better prognosis as they usually possessed more intangible (e.g., social support, access to health care) and tangible resources (e.g., medication, food supplies, housing) compared to their peers who were at lower SES [10][11][12][13].
In addition to PLWHA's prognosis, studies have highlighted the importance of addressing PLWHA's psychosocial needs and behaviors of substance use, as those with psychosocial comorbidities and problems of substance use are at a higher risk of sexual disinhibition, not adhering to ART, having lower quality of life and encountering other health problems, which might negatively influence the social and individual well-beings [14,15]. In addition to their psychosocial distress, stigmatized experience is commonly encountered in the life context of PLWHA [16,17].
Stigma was defined as a discrediting attribute of a given individual as a result of the possession of socially devalued marks [18]. In a continuous social process and complex social interactions including labeling, stereotyping, separation, status loss, and discrimination, the possessor of the socially devalued marks has been stigmatized [18,19]. Following a similar process, HIV stigma against PLWHA has been developed and maintained within different social and cultural contexts [15,20,21]. Scholars have revealed a continuum of devastating consequences (physical, emotional and financial burdens) of HIV stigma against PLWHA in China and other settings [16,17,[22][23][24].
In the current study, we aimed to examine SES stratum specific associations between different types of HIV stigma and psychosocial well-being as well as behaviors of substance use among a group of PLWHA recruited from Chinese settings. Our hypotheses are that HIV stigma is positively associated with psychological distress and problems of substance use among PLWHA, and such impacts may interact with different SES gradients.

Study site
The current study was conducted in Guangxi Zhuang Autonomous Region (Guangxi) in China. Guangxi is located in the Southwest of China with a total population of 50 million with 32% of Zhuang-ethnic and 8% of other minorities [25]. HIV epidemic has been surging in Guangxi. By the end of 2014, Guangxi has ranked first among all 31 provinces in terms of cumulative HIV seropositive cases [26,27].
in the current analysis. Specifically about 80% of the included participants completed the questionnaire assisted by the well-trained local CDC staff and health care workers. The rest of the sample completed the questionnaire on their own. Written consent was obtained from all participants. The Institutional Review Boards at Wayne State University in the United States and Guangxi CDC in China reviewed and approved the research protocol and the consent procedure.
psychosocial well-being indicators and substance use behaviors using ANOVA (for continuous variables) and Chi-squares tests (for categorical variables) among PLWHA at different SES gradients. Second, we explored how different types of HIV stigma impacted psychosocial wellbeing and substance use behaviors using multivariate regression analyses while adjusting for association-specific confounders. In addition, stratified analyses by different SES gradients (e.g., years of education, job category, residence type and monthly income) were further conducted. All data analyses were conducted using Stata 12.01 (StataCorp LP, College Station, Texas, USA).

Psychosocial wellbeing among PLWHA at different SES gradients
There were significant differences by SES gradients for most indicators of psychosocial wellbeing and substance use behaviors. Specifically, PLWHA with longer years of school, higher income levels, living at urban areas and working as "others" showed significantly better psychosocial wellbeing, less drug use behaviors, and less perceived and internalized stigma (p<0.05) ( Table 2).

Discussion
Our analyses revealed that stigma significantly impacted PLWHA's psychosocial distress and behaviors of substance use, and the impacts varied by different SES gradients. Specifically, PLWHA with higher education may be more sensitive compared with these who had lower educational attainment. As a result, the magnitude of each type of stigma was higher in the more educated group. On the other hand, PLWHA with higher income may play as a buffer from suffering stigma, therefore, they were least impacted by HIV stigma. However, the enacted stigma had the highest odds on self-esteem, smoking and alcohol use among PLWHA with the highest level of monthly income. Perhaps PLWHA with higher income were more sensitive to their stigmatized experience (e.g., enacted stigma), but less likely to generate negative opinion towards themselves (e.g., internalized and perceived stigma). However, PLWHA with higher income were more likely to report higher likelihood of substance use than their peers with lower incomes. It is perhaps that PLWHA with higher income were more likely to afford their expense of substances compared to people with lower income.
Stratified analyses by the SES gradients provided a detailed profile for PLWHA with different SES were impacted by HIV stigma with various magnitudes, although the overall analyses can provide a general direction. For instance, the overall analyses revealed enacted stigma increased the odds of having anxiety problem among PLWHA by 3.8 times after controlling for potential confounders. Our stratified analyses indicated PLWHA with lower education and least income (e.g., <999 RMB/month) were impacted to the greatest extend by enacted stigma compared to their peers. Although the overall analyses revealed non-significant odds of drinking alcohol among PLWHA who encountered enacted stigma, our stratified analyses indicated that enacted stigma increased the odds of drinking alcohol for PLWHA with monthly income more than 3,000 RMB by 2.4 times, and increased the odds for PLWHA who were selfemployed by 1.91 times. If no stratified analyses had been conducted, we may conclude that enacted stigma had least impact on PLWHA's drinking behaviors. The differences between overall and stratified analyses can be observed throughout other indicators for psychosocial status and substance use in the current study. Such findings aimed to remind health professionals of having a pair of keen eyes to observe subtle differences among PLWHA with different SES backgrounds in order to design tailored interventions.
Consistent with existing studies, the psychosocial distress and substance use behaviors were more common among individuals with lower SES or living in lower-graded neighborhoods [9,33,34]. People having higher SES usually meant they possess more material (e.g., money, house), human (e.g., skills, knowledge), and social capital (e.g., social support, social network) compared to people with lower SES [5,35]. PLWHA with higher SES usually had lower mortality rate and better prognosis than their low-SES counterparts [3]. How the SES interacted with the association between HIV stigma and psychosocial problems was complex and required a more sophisticated study design to explore the mechanisms. Our study is the first paper to document the paucity of SES stratified analyses between HIV stigma and psychosocial status among PLWHA in China.
A few caveats should be acknowledged when interpreting findings in the current study. First, the nature of cross-sectional design constrained our capacity to make any causal inferences between HIV stigma and psychosocial status as well as substance use among PLWHA. We call for longitudinal studies in future research to examine any potential causality. Second, Table 5. Associations between different types of stigma and psychosocial well-being as well as substance use by residence status (rural vs. urban) (N = 2987).

Stigma
Urban  all data were collected via self-reporting, participants may underreport their substance use or psychosocial status due to social desirability bias. Therefore, misclassification of key variables may result in diluting the effect of stigma toward the null. Although audio computer-assisted self-interview techniques may increase the validity of collected data, the evidence was inconclusive as traditional face-to-face interview may not be uniformly inferior to non-interviewer techniques across all occasions [36]. Third, as the current study was conducted in a southwest region with many minorities, findings in the current study may be subject to limited generalizability to other settings in China. Fourth, due to the dynamic changes of the definition on SES, the way of measuring the SES among PLWHA in the current study may not be able to capture all domains of individual's social, human and material capital as literatures suggested [5]. However, we followed the guideline suggested by the DHHS for practical and feasible purposes [6]. Last, due to limited space of the questionnaire, we did not collect detailed information on   participants' job categories. PLWHA who have "decent" jobs (e.g., government employees or school teachers) may encounter more HIV stigma compared to PLWHA working as peasant workers or service staff. In future studies, we suggest researchers employ hypotheses-driven strategies to explore how SES impact these studied associations. To our knowledge, it is the first study to assess the stratified impacts of HIV stigma on PLWHA's psychosocial status and behaviors of substance use by their SES gradients. Findings in the current study served as guideline for health professionals to design tailored interventions among PLWHA in future. For PLWHA with better education, living in urban areas and not working as peasant workers or unemployed, more interventions focusing on psychosocial health are needed. For PLWHA with higher income, more programs on improving coping strategies and reducing their substance use behaviors are desired. Only can we combine individual-level factors with social, historical and biophysical contexts of PLWHA, health professionals can better understand the disease etiology, health, and intervention modes for PLWHA in China.

Author Contributions
Conceptualization: XL SQ YZ.