Analysis of in vitro fertilization data with multiple outcomes using discrete time-to-event analysis

Area of Science:

• Reproductive Medicine
• Biostatistics
• Clinical Research

Background:

• In vitro fertilization (IVF) is a common assisted reproductive technology.
• IVF studies offer opportunities to identify factors influencing reproductive success.
• Analyzing complex IVF data, including multiple cycles and outcomes, presents a significant challenge.

Purpose of the Study:

• To develop advanced statistical methodologies for analyzing complex IVF data.
• To address the challenge of multiple cycles and multiple failure types in IVF studies.
• To enable the full utilization of data from IVF research.

Main Methods:

• Developed a generalized linear modeling framework for IVF data analysis.
• Incorporated various models, including shared frailty, failure-specific frailty, and transitional models.
• Applied the methodology to a real-world IVF study dataset and conducted simulation studies.

Main Results:

• The proposed statistical framework effectively handles the complexity of IVF data.
• The methodology allows for the implementation of diverse modeling approaches using standard software.
• Performance evaluation through simulation demonstrated the robustness of the developed methods.

Conclusions:

• The new statistical approach enhances the analysis of complex IVF data.
• This methodology facilitates a more comprehensive understanding of factors affecting reproductive outcomes in IVF.
• The findings support improved data utilization in assisted reproductive technology research.