Social Data Commons: Access to Care (HOI) (v2.0.0)

Overview Access to care composite index for Virginia census tracts, counties, and health districts. Combines physician availability (primary care physicians within 30 driving miles, derived from CMS Medicare Physician & Other Practitioners PUF and OSRM-based block-group travel times) with insurance coverage (ACS 5-year uninsured rate for ages 19-64) into a z-score composite where higher values indicate better access to care. This dataset is produced by the Social Data Commons at the University of Virginia as part of the Access To Care data pipeline. Provenance Based on the Access to Care methodology originally developed by the Virginia Department of Health (VDH), Office of Minority Health and Health Equity (OMHHE) as part of their Health Opportunity Index (HOI). This pipeline reproduces and extends the VDH indicator using open data sources: CMS Medicare Physician PUF (primary care providers in VA, assigned to census tracts via HUD ZIP-to-tract crosswalk with residential ratio weighting) and ACS 5-year uninsured estimates. Physician accessibility uses OSRM-based driving distances between block group centroids, aggregated to tract level, with a 30-mile threshold. The composite index is the negated z-score of the sum of z-scored population-per-physician ratio and z-scored uninsured percentage. Coverage Temporal coverage: 2017–2023 (annual, CMS PUF + ACS 5-year) Geographic levels: County, Health District, Tract Coverage areas: Virginia (statewide) Methodology Composite z-score index measuring access to primary care. Combines two components: (1) population-to-physician ratio, using CMS Medicare Physician PUF data to count primary care physicians (Internal Medicine, Family Practice, Pediatric Medicine, OB/GYN) within 30 miles of each census tract via OSRM driving distances and HUD ZIP-to-tract crosswalk allocation; and (2) percentage of uninsured residents aged 19-64 from ACS 5-year estimates (B27010). The index is computed as -1 × z(z(pop_per_physician) + z(pct_uninsured)), so higher values indicate better access to care. Source Tables Medicare Physician & Other Practitioners - by Provider, 2017-2023 ACS 5-Year Estimates, Tables B01001 (Total Population) and B27010 (Health Insurance Coverage) USPS ZIP Code-Census Tract Crosswalk, December 2021 Block-group-to-block-group driving distances, 8-state region (VA, MD, DC, DE, NC, TN, KY, WV) Variables B01001_001: Tot Pop B27010_033: Uninsured 19 34 B27010_050: Uninsured 35 64 Measures (1) Note on naming conventions: Measures containing _geo20 are computed using 2020 Census geographic boundaries. access_care_indicator_geo20: Access to care indicator (2020 geographies) (mean) Composite z-score of physician availability within 30 miles and insurance coverage rate. Data Sources CMS Medicare Physician & Other Practitioners PUF (accessed 2025) U.S. Census Bureau (accessed 2025) HUD Office of Policy Development and Research (accessed 2025) OSRM (Open Source Routing Machine) (accessed 2025) File Format Data files are provided as xz-compressed CSV (.csv.xz) with the following columns: geoid, region_type, region_name, year, measure, value, moe (margin of error, where available).概述 本数据集为弗吉尼亚州普查街区、县及卫生辖区构建的医疗可及性综合指数。该指数将医师可及性（30英里驾车范围内的初级护理医师，数据来源于CMS医疗保险医师及其他从业者公共使用文件集（CMS Medicare Physician & Other Practitioners PUF）与基于开源路由机（Open Source Routing Machine, OSRM）的普查区块组出行时间）与保险覆盖情况（美国社区调查（American Community Survey, ACS）5年期19-64岁人群无保险率）整合为Z分数（z-score）综合指数，指数值越高代表医疗可及性越好。本数据集由弗吉尼亚大学社会数据共享平台（Social Data Commons）作为“医疗可及性”数据管道的一部分生成。 数据溯源 本数据集基于弗吉尼亚州卫生署（Virginia Department of Health, VDH）少数族裔卫生与健康公平办公室（Office of Minority Health and Health Equity, OMHHE）最初为其《健康机会指数（Health Opportunity Index, HOI）》开发的医疗可及性方法构建。本数据管道使用开源数据源复刻并拓展了VDH的相关指标：包括CMS医疗保险医师公共使用文件集（用于获取弗吉尼亚州的初级护理提供者，通过住房和城市发展部（Housing and Urban Development, HUD）邮编-普查街区跨区映射表结合居住比例加权分配至各普查街区）以及ACS 5年期无保险率估算数据。医师可及性指标采用基于OSRM的普查区块组质心间驾车距离，聚合至街区层级并设置30英里阈值。综合指数为每医师人口比Z分数与无保险率Z分数之和的负Z分数。 覆盖范围 时间覆盖范围：2017–2023年（年度更新，数据来源于CMS PUF与ACS 5年期数据集） 地理层级：县、卫生辖区、普查街区 覆盖区域：弗吉尼亚州全域 方法学 本综合Z分数指数用于衡量初级医疗可及性，包含两个核心维度：(1) 每医师人口比：通过CMS医疗保险医师及其他从业者公共使用文件集数据，结合OSRM驾车距离与HUD邮编-普查街区跨区分配规则，统计各普查街区30英里范围内的初级护理医师（内科、全科医学、儿科、妇产科）数量；(2) 19-64岁无保险居民占比：来源于ACS 5年期估算数据（表B27010）。该指数计算公式为-1 × Z(Z(pop_per_physician) + Z(pct_uninsured))，因此指数值越高代表医疗可及性越好。 源数据表 1. 2017-2023年《按提供者分类的医疗保险医师及其他从业者》数据集 2. 美国社区调查（ACS）5年期估算数据，包含表B01001（总人口）与表B27010（健康保险覆盖情况） 3. 2021年12月《美国邮政总局邮编-普查街区跨区映射表》 4. 8州区域（弗吉尼亚州、马里兰州、哥伦比亚特区、特拉华州、北卡罗来纳州、田纳西州、肯塔基州、西弗吉尼亚州）普查区块组间驾车距离数据集 变量 - B01001_001：总人口（Tot Pop） - B27010_033：19-34岁无保险人群 - B27010_050：35-64岁无保险人群 指标（1） 命名规范说明：带有_geo20后缀的指标基于2020年普查地理边界计算得到。 access_care_indicator_geo20：医疗可及性指标（2020地理边界，均值）：30英里内医师可及性与保险覆盖率的综合Z分数。 数据来源 1. 美国医疗保险与医疗补助服务中心（CMS）医疗保险医师及其他从业者公共使用文件集（2025年获取） 2. 美国人口普查局（2025年获取） 3. 住房和城市发展部政策发展与研究办公室（2025年获取） 4. 开源路由机（OSRM，2025年获取） 文件格式 数据文件采用xz压缩的CSV格式（.csv.xz），包含以下字段：地理标识符（geoid）、区域类型（region_type）、区域名称（region_name）、年份（year）、指标名称（measure）、指标值（value）、边际误差（moe，如可获取）。