Scholarlink

Purpose <p>Public health surveillance systems require high-quality data to represent the population. In the absence of comprehensive or random testing throughout the COVID-19 pandemic, we have developed a proxy method for synthetic random sampling to estimate the actual community-level viral incidence, based on viral testing of patients who are asymptomatic and present for elective procedures within a hospital system.</p> Methods <p>The approach collects routine testing data on SARS-CoV-2 exposure among outpatients and performs statistical adjustments of sample representation using multilevel regression and poststratification (MRP), a procedure that adjusts for nonrepresentativeness of the sample and yields stable small group estimates. We extend MRP to accommodate time-varying data and granular geography.</p> Results <p>We have developed an open-source, user-friendly MRP interface for public implementation of the Bayesian analysis workflow. We illustrate the MRP interface with an application to track community-level COVID-19 viral transmission in Michigan. We present the estimated infection rate over time for the targeted population and across demographic and geographic subpopulations.</p> Conclusion <p>The interface provides timely, substantive insights into population health trends and serves as a valuable surveillance tool for future epidemic preparedness. Beyond monitoring COVID-19, the MRP interface can analyze a wide range of health and social science data, making it broadly applicable to diverse research areas with reproducibility and scientific rigor.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Multilevel regression and poststratification interface: an application to track community-level COVID-19 viral transmission

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月3日

查看全文
  • Yajuan Si,
  • Toan Tran,
  • Jonah Gabry,
  • Mitzi Morris,
  • Andrew Gelman

摘要

Purpose

Public health surveillance systems require high-quality data to represent the population. In the absence of comprehensive or random testing throughout the COVID-19 pandemic, we have developed a proxy method for synthetic random sampling to estimate the actual community-level viral incidence, based on viral testing of patients who are asymptomatic and present for elective procedures within a hospital system.

Methods

The approach collects routine testing data on SARS-CoV-2 exposure among outpatients and performs statistical adjustments of sample representation using multilevel regression and poststratification (MRP), a procedure that adjusts for nonrepresentativeness of the sample and yields stable small group estimates. We extend MRP to accommodate time-varying data and granular geography.

Results

We have developed an open-source, user-friendly MRP interface for public implementation of the Bayesian analysis workflow. We illustrate the MRP interface with an application to track community-level COVID-19 viral transmission in Michigan. We present the estimated infection rate over time for the targeted population and across demographic and geographic subpopulations.

Conclusion

The interface provides timely, substantive insights into population health trends and serves as a valuable surveillance tool for future epidemic preparedness. Beyond monitoring COVID-19, the MRP interface can analyze a wide range of health and social science data, making it broadly applicable to diverse research areas with reproducibility and scientific rigor.