Kenyon Chris R, Buyze Jozefien, Schwartz Ilan S
Clinical Science, Institute of Tropical Medicine, Antwerp, 2000, Belgium.
University of Alberta, Alberta, Canada.
F1000Res. 2018 Dec 2;7:1879. doi: 10.12688/f1000research.17108.1. eCollection 2018.
It is unclear why HIV prevalence varies by nearly two orders of magnitude between regions within countries in sub-Saharan Africa. In this ecological study, we assess if HIV prevalence by region is associated with any of four markers of higher risk sexual behavior: lifetime number of partners, multiple partners in past year, higher risk sex (defined as sex with non-cohabiting, non-marital partners) and age at debut. We performed Pearson's correlation between the 4 behavioral risk factors and HIV prevalence by region in 47 nationally representative surveys from 27 sub-Saharan African countries, separately by gender. In addition, principal components analysis was used to reduce the eight risk factors (four for each gender) to two principal components (PCs). Mixed effects linear regression was used to assess the relationship between the resulting two PCs and HIV prevalence after controlling for the prevalence of male circumcision. HIV prevalence varied by a median 3.7 fold (IQR 2.9-7.9) between regions within countries. HIV prevalence was strongly associated with higher risk sex and, to a lesser extent, the other risk factors evaluated. Both PCs were strongly associated with HIV prevalence when assessed via linear regression. Differences in sexual behavior may underpin the large differences in HIV-prevalence between subpopulation within sub-Saharan African countries.
目前尚不清楚为何撒哈拉以南非洲国家内部不同地区的艾滋病毒流行率相差近两个数量级。在这项生态学研究中,我们评估按地区划分的艾滋病毒流行率是否与高风险性行为的四个标志物中的任何一个相关:终身性伴侣数量、过去一年中的多个性伴侣、高风险性行为(定义为与非同居、非婚姻伴侣发生性行为)以及初次性行为的年龄。我们在来自27个撒哈拉以南非洲国家的47项具有全国代表性的调查中,按性别分别对4个行为风险因素与按地区划分的艾滋病毒流行率进行了Pearson相关性分析。此外,主成分分析用于将八个风险因素(每个性别四个)缩减为两个主成分(PCs)。在控制了男性包皮环切术的流行率后,使用混合效应线性回归来评估所得的两个主成分与艾滋病毒流行率之间的关系。国家内部不同地区的艾滋病毒流行率中位数相差3.7倍(四分位距2.9 - 7.9)。艾滋病毒流行率与高风险性行为密切相关,在较小程度上也与评估的其他风险因素相关。通过线性回归评估时,两个主成分均与艾滋病毒流行率密切相关。性行为的差异可能是撒哈拉以南非洲国家内不同亚人群之间艾滋病毒流行率存在巨大差异的原因。