Estimating the Population Sizes of Men Who Have Sex With Men in US States and Counties Using Data From the American Community Survey.
Estimating the Population Sizes of Men Who Have Sex With Men in US States and Counties Using Data From the American Community Survey.
复制标题
DOI：
10.2196/publichealth.5365
复制
发表时间：
2016-01
影响因子：
8.5
通讯作者：
Rosenberg ES
中科院分区：
文献类型：
--
作者：
Grey JA;Bernstein KT;Sullivan PS;Purcell DW;Chesson HW;Gift TL;Rosenberg ES
在美国，每年新诊断的人类免疫缺陷病毒(HIV)和相当数量的性传播感染(STI)中，男性与男性之间的性传播所占比例最大。然而，男男性行为者(MSM)中艾滋病毒和其他性传播疾病的流行率和年发病率无法在当地环境下估计，因为有关性行为，特别是同性行为的人口统计数据通常不是通过大规模调查收集的，这些调查允许在州、县或更精细的水平上进行分析，例如美国十年一次的人口普查或美国社区调查(ACS)。因此，间接估计男男性接触者人口规模的技术是必要的，以提供不同地理水平的比率的分母。我们的目标是间接估计县一级的男男性接触者人口规模，纳入最新的数据估计，并将县一级的估计汇总到各州和以核心为基础的统计区(CBSA)。我们使用美国城市协会的数据来计算美国每个县的权重，计算的基础是由男性为户主的家庭的相对比例，以及相同城市化水平的县之间的总体比例(即，大的中心大都市区县、大的边缘大都市区县、中/小大都市区县或非大都市区县)。然后，我们使用这个权重来调整过去一年中与男性发生性关系的成年男性在每个县的城市化分层百分比，这是根据国家健康和营养检查调查(NHANES)得出的估计。我们将加权百分比乘以每个县的成年男性人数，以估计其男男性接触者的数量，然后将县级估计值相加，得出州和CBSA级的估计值。最后，我们将我们估计的MSM人口规模调整为对过去5年美国MSM比例的荟萃分析估计(3.9%)。我们发现，在各州成年男性中MSM的比例从1.5%(怀俄明州)到6.0%(罗德岛州)不等。美国超过四分之一的男男性接触者居住在13个县中的1个县。在人口超过30万的县中，男男性接触者的县级比例最高的五个县是加利福尼亚州旧金山县，18.5%(66,586/359,566)；纽约县，13.8%(87,556/635,847)；科罗拉多州丹佛县，10.5%(25,465/243,002)；俄勒冈州马尔特诺马县，9.9%(28,949/292,450)；马萨诸塞州萨福克县，9.1%(26,338/289,634)。虽然加利福尼亚州(n=792,750)和洛杉矶县(n=251,521)分别拥有最多的州和县的MSM人口，但纽约市-纽瓦克-泽西市CBSA的MSM人数是所有CBSA中最多的(n=397,399)。我们使用了一种新的方法来生成MSM人口的小区域估计，结合了先前的工作、最近的数据和特定于城市化的参数。我们还使用了推算方法来估计农村地区的男男性接触者，在那里同性性行为可能被低估。我们的方法得出了美国各州、县和大都市区内男男性接触者人口规模的估计，这为计算这些地理水平上的艾滋病毒和性传播感染的流行率和发病率提供了分母。
In the United States, male-to-male sexual transmission accounts for the greatest number of new human immunodeficiency virus (HIV) diagnoses and a substantial number of sexually transmitted infections (STI) annually. However, the prevalence and annual incidence of HIV and other STIs among men who have sex with men (MSM) cannot be estimated in local contexts because demographic data on sexual behavior, particularly same-sex behavior, are not routinely collected by large-scale surveys that allow analysis at state, county, or finer levels, such as the US decennial census or the American Community Survey (ACS). Therefore, techniques for indirectly estimating population sizes of MSM are necessary to supply denominators for rates at various geographic levels. Our objectives were to indirectly estimate MSM population sizes at the county level to incorporate recent data estimates and to aggregate county-level estimates to states and core-based statistical areas (CBSAs). We used data from the ACS to calculate a weight for each county in the United States based on its relative proportion of households that were headed by a male who lived with a male partner, compared with the overall proportion among counties at the same level of urbanicity (ie, large central metropolitan county, large fringe metropolitan county, medium/small metropolitan county, or nonmetropolitan county). We then used this weight to adjust the urbanicity-stratified percentage of adult men who had sex with a man in the past year, according to estimates derived from the National Health and Nutrition Examination Survey (NHANES), for each county. We multiplied the weighted percentages by the number of adult men in each county to estimate its number of MSM, summing county-level estimates to create state- and CBSA-level estimates. Finally, we scaled our estimated MSM population sizes to a meta-analytic estimate of the percentage of US MSM in the past 5 years (3.9%). We found that the percentage of MSM among adult men ranged from 1.5% (Wyoming) to 6.0% (Rhode Island) among states. Over one-quarter of MSM in the United States resided in 1 of 13 counties. Among counties with over 300,000 residents, the five highest county-level percentages of MSM were San Francisco County, California at 18.5% (66,586/359,566); New York County, New York at 13.8% (87,556/635,847); Denver County, Colorado at 10.5% (25,465/243,002); Multnomah County, Oregon at 9.9% (28,949/292,450); and Suffolk County, Massachusetts at 9.1% (26,338/289,634). Although California (n=792,750) and Los Angeles County (n=251,521) had the largest MSM populations of states and counties, respectively, the New York City-Newark-Jersey City CBSA had the most MSM of all CBSAs (n=397,399). We used a new method to generate small-area estimates of MSM populations, incorporating prior work, recent data, and urbanicity-specific parameters. We also used an imputation approach to estimate MSM in rural areas, where same-sex sexual behavior may be underreported. Our approach yielded estimates of MSM population sizes within states, counties, and metropolitan areas in the United States, which provide denominators for calculation of HIV and STI prevalence and incidence at those geographic levels.