This chapter discusses the statistical analysis of recurrent events within the estimand framework. Recurrent events are those that can occur repeatedly for a single patient, such as heart failure hospitalizations, relapses in multiple sclerosis, or exacerbations in chronic obstructive pulmonary disease (COPD) or asthma. Focusing on a two-arm clinical trial scenario where discontinuation of the randomized treatment for any reason is the only intercurrent event, this chapter illustrates how to define the corresponding estimand attributes when addressing the intercurrent event using one of the five strategies outlined in the ICH E9(R1) guideline. For each estimand, appropriate estimation methods are presented. The methods are illustrated with a case study using synthetic data from a clinical trial in asthma. The chapter concludes with a review of the literature on estimands for recurrent events in the presence of death and remarks on software solutions for the presented estimation approaches.

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

Statistical Analysis Under the Estimand Framework: Recurrent Events

  • Tobias Mütze,
  • Tianmeng Lyu

摘要

This chapter discusses the statistical analysis of recurrent events within the estimand framework. Recurrent events are those that can occur repeatedly for a single patient, such as heart failure hospitalizations, relapses in multiple sclerosis, or exacerbations in chronic obstructive pulmonary disease (COPD) or asthma. Focusing on a two-arm clinical trial scenario where discontinuation of the randomized treatment for any reason is the only intercurrent event, this chapter illustrates how to define the corresponding estimand attributes when addressing the intercurrent event using one of the five strategies outlined in the ICH E9(R1) guideline. For each estimand, appropriate estimation methods are presented. The methods are illustrated with a case study using synthetic data from a clinical trial in asthma. The chapter concludes with a review of the literature on estimands for recurrent events in the presence of death and remarks on software solutions for the presented estimation approaches.