Additive rates model for recurrent event data with intermittently observed time-dependent covariates

Area of Science:

• Biostatistics
• Epidemiology

Background:

• Recurrent event data analysis is crucial in many fields, including epidemiology.
• Semiparametric additive rates models offer interpretable regression coefficients for event rates.
• Accurate estimation requires continuous time-dependent covariate data, which is often unavailable.

Purpose of the Study:

• To develop a robust statistical method for analyzing recurrent event data when time-dependent covariates are measured intermittently.
• To address the limitations of existing methods that require complete covariate data.
• To provide a practical approach for epidemiological studies with sparse covariate measurements.

Main Methods:

• Proposed a novel kernel smoothing technique applied to functions of time-dependent covariates within the estimating function.
• Avoided individual covariate trajectory imputation, focusing on cross-subject smoothing.
• Evaluated the method's performance through simulation studies.

Main Results:

• The proposed kernel smoothing method demonstrated superior performance over simple imputation techniques in simulations.
• The method effectively handles intermittently measured time-dependent covariates in recurrent event data analysis.
• The approach was successfully applied to an epidemiologic dataset on recurrent pharyngitis.

Conclusions:

• Kernel smoothing offers a viable and effective solution for analyzing recurrent event data with intermittently measured covariates.
• This method enhances the applicability of additive rates models in real-world epidemiological research.
• The findings support the use of this technique for more accurate analysis of recurrent health events.