[Using capture-recapture models to estimate transition rates between states in interval-censored data].

Area of Science:

• Epidemiology
• Biostatistics
• Clinical Research

Context:

• Longitudinal studies in clinical research often suffer from imperfect patient follow-up, leading to interval-censored data.
• Estimating survival or transition rates between disease stages is crucial for understanding disease progression and treatment efficacy.
• Traditional methods may yield biased estimates due to unmodelled patient compliance.

Purpose:

• To adapt capture-recapture models from population biology for epidemiological and clinical research.
• To estimate disease stage transition rates in patients with interval-censored data due to imperfect follow-up.
• To provide unbiased survival estimates by jointly modeling patient compliance.

Summary:

• Capture-recapture models, typically used in ecology, are introduced as a novel method for epidemiological studies.
• The models estimate survival and transition rates between disease stages, even with interval-censored data from longitudinal studies with imperfect follow-up.
• Unbiased estimates are achieved by simultaneously modeling patient compliance, defined as the probability of adhering to scheduled visits.

Impact:

• The adapted capture-recapture methodology offers a robust approach for analyzing complex clinical trial data.
• Application to cancer data reveals that time elapsed since study entry significantly influences patient compliance.
• This work enhances the accuracy of survival and transition rate estimations in disease progression research.