Analysis of Inter-Arrival Times in Presence of Panel Count Data with Intermittent Examination Times: An Application to Spontaneous Labor in Women

• Activity Number: 641
• Type: Topic Contributed
• Date/Time: Thursday, August 13, 2015 : 8:30 AM to 10:20 AM
• Sponsor: ENAR
• Author(s): Rajeshwari Sundaram* and Ling Ma
• Companies: NIH and Eunice Kennedy Shriver National Institute of Child Health and Human Development
• Keywords: Survival analysis ; informative censoring ; recurrent events ; epidemiological studies

Abstract

In longitudinal studies of serial events where each subject may be observed only at several distinct and random observation times, only the numbers of occurrences of the events are known at the observation times leading to panel count data. Most of the existing methods for panel count data focus on statistical inference for the point process while it is also of interest to make inference on the inter-arrival times. The application of interest in this project is to provide a framework for modeling gap-time distributions between cervical dilations in the first stage labor process (e.g. 3 cm dilation to 4 cm dilation). One well-known problem in obstetrics is that the start of labor is not clearly defined. We propose a flexible semi-parametric model for the panel count data and observation process while accounting for the unmeasured time, i.e., time from start of dilation to admit time. We propose a likelihood based approach to estimate various quantities of interest. The proposed approach is investigated through finite sample properties and the large sample properties of the proposed estimators. We illustrate the proposed approach through Collaborative Perinatal Project study data.