Abstract：
OBJECTIVE: Public Health officials are often challenged to effectively allocate limited resources. Social determinants of health (SDOH) may cluster in areas to cause unique profiles related to various adverse life events. The authors use the framework of unintended teen pregnancies to illustrate how to identify the most vulnerable neighborhoods.
METHODS: This study used data from the U.S. American Community Survey, Princeton Eviction Lab, and Connecticut Office of Vital Records. Census tracts are small statistical subdivisions of a county. Latent class analysis (LCA) was employed to separate the 832 Connecticut census tracts into four distinct latent classes based on SDOH, and GIS mapping was utilized to visualize the distribution of the most vulnerable neighborhoods. GEE Poisson regression model was used to assess whether latent classes were related to the outcome. Data were analyzed in May 2021.
RESULTS: LCA\'s results showed that class 1 (non-minority non-disadvantaged tracts) had the least diversity and lowest poverty of the four classes. Compared to class 1, class 2 (minority non-disadvantaged tracts) had more households with no health insurance and with single parents; and class 3 (non-minority disadvantaged tracts) had more households with no vehicle available, that had moved from another place in the past year, were low income, and living in renter-occupied housing. Class 4 (minority disadvantaged tracts) had the lowest socioeconomic characteristics.
CONCLUSIONS: LCA can identify unique profiles for neighborhoods vulnerable to adverse events, setting up the potential for differential intervention strategies for communities with varying risk profiles. Our approach may be generalizable to other areas or other programs.
CONCLUSIONS: What is already known on this topic Public health practitioners struggle to develop interventions that are universally effective. The teen birth rates vary tremendously by race and ethnicity. Unplanned teen pregnancy rates are related to multiple social determinants and behaviors. Latent class analysis has been applied successfully to address public health problems. What this study adds While it is the pregnancy that is not planned rather than the birth, access to pregnancy intention data is not available resulting in a dependency on teen birth data for developing public health strategies. Using teen birth rates to identify at-risk neighborhoods will not directly represent the teens at risk for pregnancy but rather those who delivered a live birth. Since teen birth rates often fluctuate due to small numbers, especially for small neighborhoods, LCA may avoid some of the limitations associated with direct rate comparisons. The authors illustrate how practitioners can use publicly available SDOH from the Census Bureau to identify distinct SDOH profiles for teen births at the census tract level. How this study might affect research, practice or policy These profiles of classes that are at heightened risk potentially can be used to tailor intervention plans for reducing unintended teen pregnancy. The approach may be adapted to other programs and other states to prioritize the allocation of limited resources.
METHODS: This study used data from the U.S. American Community Survey, Princeton Eviction Lab, and Connecticut Office of Vital Records. Census tracts are small statistical subdivisions of a county. Latent class analysis (LCA) was employed to separate the 832 Connecticut census tracts into four distinct latent classes based on SDOH, and GIS mapping was utilized to visualize the distribution of the most vulnerable neighborhoods. GEE Poisson regression model was used to assess whether latent classes were related to the outcome. Data were analyzed in May 2021.
RESULTS: LCA\'s results showed that class 1 (non-minority non-disadvantaged tracts) had the least diversity and lowest poverty of the four classes. Compared to class 1, class 2 (minority non-disadvantaged tracts) had more households with no health insurance and with single parents; and class 3 (non-minority disadvantaged tracts) had more households with no vehicle available, that had moved from another place in the past year, were low income, and living in renter-occupied housing. Class 4 (minority disadvantaged tracts) had the lowest socioeconomic characteristics.
CONCLUSIONS: LCA can identify unique profiles for neighborhoods vulnerable to adverse events, setting up the potential for differential intervention strategies for communities with varying risk profiles. Our approach may be generalizable to other areas or other programs.
CONCLUSIONS: What is already known on this topic Public health practitioners struggle to develop interventions that are universally effective. The teen birth rates vary tremendously by race and ethnicity. Unplanned teen pregnancy rates are related to multiple social determinants and behaviors. Latent class analysis has been applied successfully to address public health problems. What this study adds While it is the pregnancy that is not planned rather than the birth, access to pregnancy intention data is not available resulting in a dependency on teen birth data for developing public health strategies. Using teen birth rates to identify at-risk neighborhoods will not directly represent the teens at risk for pregnancy but rather those who delivered a live birth. Since teen birth rates often fluctuate due to small numbers, especially for small neighborhoods, LCA may avoid some of the limitations associated with direct rate comparisons. The authors illustrate how practitioners can use publicly available SDOH from the Census Bureau to identify distinct SDOH profiles for teen births at the census tract level. How this study might affect research, practice or policy These profiles of classes that are at heightened risk potentially can be used to tailor intervention plans for reducing unintended teen pregnancy. The approach may be adapted to other programs and other states to prioritize the allocation of limited resources.
摘要：
目的：公共卫生官员经常面临有效分配有限资源的挑战。健康的社会决定因素（SDOH）可能集中在区域中，以引起与各种不良生活事件相关的独特概况。作者使用意外青少年怀孕的框架来说明如何识别最脆弱的社区。
方法：这项研究使用了美国社区调查的数据，普林斯顿驱逐实验室,和康涅狄格州生命记录办公室。人口普查是一个县的小型统计分区。采用潜在类别分析（LCA）将康涅狄格州的832个人口普查区分为基于SDOH的四个不同的潜在类别，并利用GIS制图来可视化最脆弱社区的分布。GEE泊松回归模型用于评估潜在类别是否与结果相关。数据在2021年5月进行了分析。
结果：LCA\的结果表明，在四个类别中，第1类（非少数非弱势群体）的多样性最小，贫困程度最低。与第1类相比，第2类(少数群体非弱势群体)有更多没有医疗保险和单亲父母的家庭；第3类(非少数群体弱势群体)有更多没有车辆的家庭，在过去的一年里从另一个地方搬来的，低收入，住在租房者居住的住房中。第4类（少数弱势群体）的社会经济特征最低。
结论：LCA可以识别易受不良事件影响的社区的独特特征，为具有不同风险特征的社区建立不同干预策略的潜力。我们的方法可以推广到其他领域或其他程序。
结论：关于这一主题的已知公共卫生从业者正在努力开发普遍有效的干预措施。青少年的出生率因种族和种族而异。计划外青少年怀孕率与多种社会决定因素和行为有关。潜在类别分析已成功应用于解决公共卫生问题。这项研究补充了什么，虽然不是计划的怀孕而不是分娩，无法获得怀孕意向数据,导致在制定公共卫生策略时依赖青少年出生数据.使用青少年出生率来识别高危社区不会直接代表有怀孕风险的青少年，而是那些分娩活产的青少年。由于青少年出生率经常因人数少而波动，尤其是小社区，LCA可以避免与直接速率比较相关的一些限制。作者说明了从业人员如何使用人口普查局公开提供的SDOH来识别人口普查区一级青少年出生的不同SDOH概况。这项研究如何影响研究，实践或政策这些潜在风险较高的课程简介可用于制定干预计划，以减少意外的青少年怀孕。该方法可以适用于其他程序和其他州,以优先分配有限的资源。
方法：这项研究使用了美国社区调查的数据，普林斯顿驱逐实验室,和康涅狄格州生命记录办公室。人口普查是一个县的小型统计分区。采用潜在类别分析（LCA）将康涅狄格州的832个人口普查区分为基于SDOH的四个不同的潜在类别，并利用GIS制图来可视化最脆弱社区的分布。GEE泊松回归模型用于评估潜在类别是否与结果相关。数据在2021年5月进行了分析。
结果：LCA\的结果表明，在四个类别中，第1类（非少数非弱势群体）的多样性最小，贫困程度最低。与第1类相比，第2类(少数群体非弱势群体)有更多没有医疗保险和单亲父母的家庭；第3类(非少数群体弱势群体)有更多没有车辆的家庭，在过去的一年里从另一个地方搬来的，低收入，住在租房者居住的住房中。第4类（少数弱势群体）的社会经济特征最低。
结论：LCA可以识别易受不良事件影响的社区的独特特征，为具有不同风险特征的社区建立不同干预策略的潜力。我们的方法可以推广到其他领域或其他程序。
结论：关于这一主题的已知公共卫生从业者正在努力开发普遍有效的干预措施。青少年的出生率因种族和种族而异。计划外青少年怀孕率与多种社会决定因素和行为有关。潜在类别分析已成功应用于解决公共卫生问题。这项研究补充了什么，虽然不是计划的怀孕而不是分娩，无法获得怀孕意向数据,导致在制定公共卫生策略时依赖青少年出生数据.使用青少年出生率来识别高危社区不会直接代表有怀孕风险的青少年，而是那些分娩活产的青少年。由于青少年出生率经常因人数少而波动，尤其是小社区，LCA可以避免与直接速率比较相关的一些限制。作者说明了从业人员如何使用人口普查局公开提供的SDOH来识别人口普查区一级青少年出生的不同SDOH概况。这项研究如何影响研究，实践或政策这些潜在风险较高的课程简介可用于制定干预计划，以减少意外的青少年怀孕。该方法可以适用于其他程序和其他州,以优先分配有限的资源。
