Health & Wellness Attributes

Health & Wellness Attributes contains individual or household level data variables that provide and predict a variety of unique characteristics related to health ranging from BMI and health-related interests to exercise level and social determinants of health and more.

Geographic coverage: United States of America

Data Population Level: Individual or Household

Number of individuals/households covered: 242.5+ million individuals & 117+ million households

Data Source(s): Our data is created in an offline process but leverage offline and online data and behaviors. The vast majority of our database is proprietary although a few publicly available data sources are leveraged in its development.

AnalyticsIQ sources data from over 100 sources. These are predominantly public sources including: Core Demographic data from multiple sources; Census Block and Block Group level data; Econometric data from the US government; Summarized credit data from multiple credit bureaus; Property and mortgage information from county courthouses; Occupation information from state licensing boards; Past purchase behavior from catalogers and retailers that contribute their data at a category level.

AnalyticsIQ is not an original compiler as the data above is readily available for purchase out in the market. However, AnalyticsIQ uses superior analytics to make our data best-in-class.

One tool that is completely unique to AnalyticsIQ’s product development process is our proprietary survey data. This is where our Cognitive Sciences Department carefully crafts questions that we serve to a panel of consumers. Survey responses are not directly published on our file, but rather the answers are then modeled across our entire consumer file to create truly unique data points not available anywhere.

Examples of key data points include:

- Covid19_Vaccine: Likely to take COVID-19 vaccine once available
- LT_Exercise: Most likely frequency of exercise
- HW_BMI: Predicted BMI
- HW_Junk_Diet: Tendency to have an unhealthy diet
- HW_Primary_Care_Visits: Likely to have visited primary care doctor in last 12-months
- HW_Sleep_v3: Likely sleep quality
- HW_WebMD: Likely to use WebMD

For users who wish to avoid PII, all AnalyticsIQ data can be anonymized through tokenization thanks to our strong partnership with Datavant, a leading providers of data de-identification services.

健康与福祉属性数据集包含个体或家庭层面的数据变量，可提供并预测与健康相关的多种独特特征，涵盖身体质量指数（BMI）、健康相关兴趣、运动水平以及健康社会决定因素等诸多领域。

地理覆盖范围：美利坚合众国

数据覆盖层级：个体或家庭

覆盖规模：超2.425亿个体及1.17亿户家庭

数据来源说明：本数据集采用离线流程生成，同时整合离线与在线数据及用户行为数据。尽管开发过程中会使用部分公开数据源，但本数据库的绝大多数内容均为专有资产。

AnalyticsIQ 从超100个渠道获取数据，其中绝大多数为公开数据源，具体包括：多渠道核心人口统计数据、人口普查街区（Census Block）与街区组（Block Group）层级数据、美国政府发布的计量经济数据、多家征信机构汇总的信贷数据、县法院出具的房产与抵押信息、州执业监管委员会提供的职业信息，以及分类目录商与零售商提供的品类级过往购买行为数据。

AnalyticsIQ 并非原始数据汇编方，因为上述数据均可在市场上直接采购获取。但本机构凭借领先的数据分析技术，打造出行业顶尖水准的数据集。

本机构产品开发流程中独有的一项工具为专有调研数据：我们的认知科学部门精心设计问卷，并面向消费者调研面板投放。调研结果不会直接录入数据集，而是通过建模将应答信息投射至全量消费者数据库中，由此生成全球独有的数据特征。

核心数据点示例如下：
- Covid19_Vaccine：新冠疫苗获批后接种意愿
- LT_Exercise：预期运动频率
- HW_BMI：预测身体质量指数（BMI）
- HW_Junk_Diet：不健康饮食倾向
- HW_Primary_Care_Visits：过去12个月内就诊全科医生的可能性
- HW_Sleep_v3：预期睡眠质量
- HW_WebMD：使用WebMD医疗平台的意愿

对于希望避免接触个人可识别信息（PII）的用户，本机构与数据去识别服务领域的领先提供商Datavant达成深度合作，可通过标记化（Tokenization）技术对所有AnalyticsIQ数据集进行匿名化处理。