PDF(Pubmed)
Abstract：
OBJECTIVE: Outbreaks of injection drug use (IDU)-associated infections have become major public health concerns in the era of the opioid epidemic. This study aimed to (1) identify county-level characteristics associated with acute HCV infection and newly diagnosed IDU-associated HIV in Oklahoma and (2) develop a vulnerability index using these metrics.
METHODS: This study employs a county-level ecological design to examine those diagnosed with acute or chronic HCV or newly diagnosed IDU-associated HIV. Poisson regression was used to estimate the association between indicators and the number of new infections in each county. Primary outcomes were acute HCV and newly diagnosed IDU-associated HIV. A sensitivity analysis included all HCV (acute and chronic) cases. Three models were run using variations of these outcomes. Stepwise backward Poisson regression predicted new infection rates and 95% confidence intervals for each county from the final multivariable model, which served as the metric for vulnerability scores.
RESULTS: Predictors for HIV-IDU cases and acute HCV cases differed. The percentage of the county population aged 18-24 years with less than a high school education and population density were predictive of new HIV-IDU cases, whereas the percentage of the population that was male, white, Pacific Islander, two or more races, and people aged 18-24 years with less than a high school education were predictors of acute HCV infection. Counties with the highest predicted rates of HIV-IDU tended to be located in central Oklahoma and have higher population density than the counties with the highest predicted rates of acute HCV infection.
CONCLUSIONS: There is high variability in county-level factors predictive of new IDU-associated HIV infection and acute HCV infection, suggesting that different public health interventions need to be tailored to these two case populations.
METHODS: This study employs a county-level ecological design to examine those diagnosed with acute or chronic HCV or newly diagnosed IDU-associated HIV. Poisson regression was used to estimate the association between indicators and the number of new infections in each county. Primary outcomes were acute HCV and newly diagnosed IDU-associated HIV. A sensitivity analysis included all HCV (acute and chronic) cases. Three models were run using variations of these outcomes. Stepwise backward Poisson regression predicted new infection rates and 95% confidence intervals for each county from the final multivariable model, which served as the metric for vulnerability scores.
RESULTS: Predictors for HIV-IDU cases and acute HCV cases differed. The percentage of the county population aged 18-24 years with less than a high school education and population density were predictive of new HIV-IDU cases, whereas the percentage of the population that was male, white, Pacific Islander, two or more races, and people aged 18-24 years with less than a high school education were predictors of acute HCV infection. Counties with the highest predicted rates of HIV-IDU tended to be located in central Oklahoma and have higher population density than the counties with the highest predicted rates of acute HCV infection.
CONCLUSIONS: There is high variability in county-level factors predictive of new IDU-associated HIV infection and acute HCV infection, suggesting that different public health interventions need to be tailored to these two case populations.
摘要：
目的：在阿片类药物流行时代，注射用药（IDU）相关感染的爆发已成为主要的公共卫生问题。这项研究旨在（1）确定与俄克拉荷马州急性HCV感染和新诊断的IDU相关HIV相关的县级特征，以及（2）使用这些指标开发脆弱性指数。
方法：本研究采用县级生态设计来检查被诊断患有急性或慢性HCV或新诊断的IDU相关HIV的患者。使用泊松回归来估计指标与每个县的新感染数量之间的关联。主要结果是急性HCV和新诊断的IDU相关HIV。敏感性分析包括所有HCV（急性和慢性）病例。使用这些结果的变化运行三个模型。逐步向后泊松回归从最终的多变量模型中预测每个县的新感染率和95％置信区间，作为漏洞得分的度量。
结果：HIV-IDU病例和急性HCV病例的预测因子不同。18-24岁的县人口中高中以下文化程度和人口密度的百分比预测了新的HIV-IDU病例，而男性人口的百分比，白色,太平洋岛民,两个或两个以上的种族，18-24岁、高中以下文化程度的人群是急性HCV感染的预测因素。HIV-IDU预测率最高的县往往位于俄克拉荷马州中部，人口密度高于急性HCV感染率预测率最高的县。
结论：预测新的IDU相关HIV感染和急性HCV感染的县级因素存在高度变异性，这表明不同的公共卫生干预措施需要针对这两个病例人群进行调整。
方法：本研究采用县级生态设计来检查被诊断患有急性或慢性HCV或新诊断的IDU相关HIV的患者。使用泊松回归来估计指标与每个县的新感染数量之间的关联。主要结果是急性HCV和新诊断的IDU相关HIV。敏感性分析包括所有HCV（急性和慢性）病例。使用这些结果的变化运行三个模型。逐步向后泊松回归从最终的多变量模型中预测每个县的新感染率和95％置信区间，作为漏洞得分的度量。
结果：HIV-IDU病例和急性HCV病例的预测因子不同。18-24岁的县人口中高中以下文化程度和人口密度的百分比预测了新的HIV-IDU病例，而男性人口的百分比，白色,太平洋岛民,两个或两个以上的种族，18-24岁、高中以下文化程度的人群是急性HCV感染的预测因素。HIV-IDU预测率最高的县往往位于俄克拉荷马州中部，人口密度高于急性HCV感染率预测率最高的县。
结论：预测新的IDU相关HIV感染和急性HCV感染的县级因素存在高度变异性，这表明不同的公共卫生干预措施需要针对这两个病例人群进行调整。
