Lim Travis W, Frangakis Constantine, Latkin Carl, Ha Tran Viet, Minh Nguyen Le, Zelaya Carla, Quan Vu Minh, Go Vivian F
Johns Hopkins Bloomberg School of Public Health, Department of International Health, Baltimore, Maryland, United States.
Johns Hopkins Bloomberg School of Public Health, Department of Biostatistics, Baltimore, Maryland, United States.
PLoS One. 2014 Mar 11;9(3):e90723. doi: 10.1371/journal.pone.0090723. eCollection 2014.
Socioeconomic status has a robust positive relationship with several health outcomes at the individual and population levels, but in the case of HIV prevalence, income inequality may be a better predictor than absolute level of income. Most studies showing a relationship between income inequality and HIV have used entire countries as the unit of analysis. In this study, we examine the association between income inequality at the community level and HIV prevalence in a sample of persons who inject drugs (PWID) in a concentrated epidemic setting. We recruited PWID and non-PWID community participants in Thai Nguyen, Vietnam, and administered a cross-sectional questionnaire; PWID were tested for HIV. We used ecologic regression to model HIV burden in our PWID study population on GINI indices of inequality calculated from total reported incomes of non-PWID community members in each commune. We also modeled HIV burden on interaction terms between GINI index and median commune income, and finally used a multi-level model to control for community level inequality and individual level income. HIV burden among PWID was significantly correlated with the commune GINI coefficient (r = 0.53, p = 0.002). HIV burden was also associated with GINI coefficient (β = 0.082, p = 0.008) and with median commune income (β = -0.018, p = 0.023) in ecological regression. In the multi-level model, higher GINI coefficient at the community level was associated with higher odds of individual HIV infection in PWID (OR = 1.46 per 0.01, p = 0.003) while higher personal income was associated with reduced odds of infection (OR = 0.98 per $10, p = 0.022). This study demonstrates a context where income inequality is associated with HIV prevalence at the community level in a concentrated epidemic. It further suggests that community level socioeconomic factors, both contextual and compositional, could be indirect determinants of HIV infection in PWID.
社会经济地位在个体和群体层面与多种健康结果有着稳固的正向关系，但就艾滋病毒流行情况而言，收入不平等可能比绝对收入水平更能预测。大多数显示收入不平等与艾滋病毒之间存在关系的研究都将整个国家作为分析单位。在本研究中，我们在一个集中流行环境下，考察社区层面的收入不平等与注射吸毒者（PWID）样本中艾滋病毒流行率之间的关联。我们在越南谅山招募了注射吸毒者和非注射吸毒者社区参与者，并进行了横断面问卷调查；对注射吸毒者进行了艾滋病毒检测。我们使用生态回归模型，根据每个公社非注射吸毒者社区成员报告的总收入计算出的基尼不平等指数，来分析我们注射吸毒者研究人群中的艾滋病毒负担。我们还根据基尼指数与公社收入中位数的交互项建立了艾滋病毒负担模型，最后使用多层次模型来控制社区层面的不平等和个体层面的收入。注射吸毒者中的艾滋病毒负担与公社基尼系数显著相关（r = 0.53，p = 0.002）。在生态回归中，艾滋病毒负担也与基尼系数（β = 0.082，p = 0.008）以及公社收入中位数（β = -0.018，p = 0.023）相关。在多层次模型中，社区层面较高的基尼系数与注射吸毒者个体感染艾滋病毒的较高几率相关（每0.01单位的优势比为1.46，p = 0.003），而较高的个人收入与较低的感染几率相关（每10美元的优势比为0.98，p = 0.022）。本研究表明，在集中流行环境下，收入不平等在社区层面与艾滋病毒流行率相关。它进一步表明，社区层面的社会经济因素，无论是背景因素还是构成因素，都可能是注射吸毒者艾滋病毒感染的间接决定因素。
