Scholarlink

This chapter explores the application of Generalized Linear Models (GLMs) in modeling binary outcomes within the context of infection-related diseases following hospital procedures. Using data from the All of Us Research Program, we compare frequentist and Bayesian logistic regression approaches, integrating both non-informative and informative priors. Furthermore, we introduce penalized regression techniques such as Ridge and LASSO and their Bayesian counterparts with shrinkage priors to address multicollinearity and high dimensionality. Our findings highlight the robustness of endocrine and respiratory disorders, as well as intubation and transplant procedures, as strong predictors of post-procedural infection. This chapter underscores the flexibility and interpretability of modern regression techniques in real-world epidemiological research.

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

Generalized Linear Models in Infectious Disease Analysis and Surveillance: Methods for Independent Data

  • 章节

  • 出版模式：

    开放获取

  • 发表日期：

    2026年

查看全文
  • Noor Muhammad Khan,
  • Ileana Baldi,
  • Maria Vittoria Chiaruttini,
  • Dario Gregori

摘要

This chapter explores the application of Generalized Linear Models (GLMs) in modeling binary outcomes within the context of infection-related diseases following hospital procedures. Using data from the All of Us Research Program, we compare frequentist and Bayesian logistic regression approaches, integrating both non-informative and informative priors. Furthermore, we introduce penalized regression techniques such as Ridge and LASSO and their Bayesian counterparts with shrinkage priors to address multicollinearity and high dimensionality. Our findings highlight the robustness of endocrine and respiratory disorders, as well as intubation and transplant procedures, as strong predictors of post-procedural infection. This chapter underscores the flexibility and interpretability of modern regression techniques in real-world epidemiological research.