Competing Risks Model for Cross-Sectional Sampled Length Biased Data

• Activity Number: 641
• Type: Topic Contributed
• Date/Time: Thursday, August 13, 2015 : 8:30 AM to 10:20 AM
• Sponsor: ENAR
• Author(s): Alexander McLain*
• Companies: University of South Carolina
• Keywords: Competing risks ; Survival analysis ; Length-biased data ; Cross-sectional studies

Abstract

Current duration data arise in cross-sectional studies from questions on the length of time from an initiating event to the time of interview. For example in the National Survey on Family Growth (NSFG), women who were considered at risk for pregnancy were asked how long they had been attempting pregnancy. Recent biostatistical methodology, uses the current lengths of pregnancy attempt to make inference on the unobserved total duration of pregnancy attempt. However, methodological gaps in these methods remain such as how to handle women that end their natural pregnancy attempt in favor of fertility treatments before they are sampled. In current methods these women must be dropped from the analysis or an assumption on independence must be made. In this paper, we develop statistical methods that can be used to estimate the distribution of the length of natural pregnancy attempts from cross-sectional data while correctly accounting for women that sought fertility treatment prior to being sampled. This amounts to a competing risks model for cross-sectional sampled length biased data. We demonstrate our approach based on simulated data, and the 2002 version of the NSFG.