Nonparametric estimation of the cumulative incidence function for doubly-truncated and interval-censored competing...

Area of Science:

• Biostatistics
• Epidemiology
• Survival Analysis

Background:

• Interval sampling is common for disease registry data.
• This sampling can lead to doubly truncated and interval-censored (DTIC) data.
• Competing risks analysis is crucial for understanding disease progression.

Purpose of the Study:

• To develop nonparametric estimators for cumulative incidence functions (CIF) with doubly-truncated and interval-censored competing risks (DTIC-C) data.
• To address challenges posed by interval sampling in disease registries.
• To provide a robust method for analyzing complex survival data.

Main Methods:

• Utilized the nonparametric maximum likelihood estimator (NPMLE) approach.
• Adapted Shen's method (Stat Methods Med Res 31:1157-1170, 2022b) for DTIC-C data.
• Developed and established consistency of novel nonparametric CIF estimators.

Main Results:

• Successfully obtained nonparametric estimators for CIF with DTIC-C data.
• Established the consistency of the proposed estimators.
• Simulation studies demonstrated good performance with finite sample sizes.

Conclusions:

• The proposed nonparametric estimators are effective for DTIC-C data from interval sampling.
• This method offers a valuable tool for disease registry analysis.
• The estimators show promise for real-world epidemiological studies.