|
Activity Number:
|
361
- SPEED: Biometrics - Methods and Application, Part 2
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 30, 2019 : 11:35 AM to 12:20 PM
|
|
Sponsor:
|
Biometrics Section
|
|
Abstract #307774
|
|
|
Title:
|
Survey Calibration to Improve the Efficiency of Pure Risk Estimates from Case-Control Samples Nested in a Cohort
|
|
Author(s):
|
Yei Eun Shin* and Ruth Pfeiffer and Barry Graubard and Mitchell Henry Gail
|
|
Companies:
|
National Cancer Institute and National Cancer Institute and National Cancer Institute and National Cancer Institute, Division of Cancer Epidemiology and Genetics
|
|
Keywords:
|
Cox model;
Inclusion probability weights;
Influence functions;
Nested case-control designs;
Pure risks;
Survey calibration
|
|
Abstract:
|
Cohort studies provide information on relative and pure risks of disease. For rare outcomes, large cohorts are needed to have sufficient numbers of events, making it costly to obtain covariate information on all members. We focus on nested case-control (NCC) designs with which Cox model can be fitted to estimate relative risks. Langholz and Borgan (1997) showed pure risks can also be estimated from NCC data. However, these approaches do not take advantage of some covariates that may be available on all cohort members. Breslow et al. (2009) improved the precision of relative risk estimates in case-cohort designs by calibrating sampling weights against imputed influences. Our objective is to extend survey calibration to general NCC designs to improve precision of estimates of relative and pure risks. We show that calibrating Samuelsen (1997) inclusion weights against 'imputed influences' and additionally 'number at risk' improves pure risk estimates. We develop explicit variance estimators for the calibrated estimates of relative and pure risk. Simulations show how much precision is improved by calibration and confirm the validity of inference based on asymptotic normality.
|
Authors who are presenting talks have a * after their name.
