Nationally Representative Absolute Risk Estimation Combining Individual Data from Epidemiologic Studies and Population-Based Surveys with Summary Statistics from Disease Registry

• Activity Number: 139 - Improving Population Inference Using Statistical Data Integration
• Type: Topic Contributed
• Date/Time: Monday, August 8, 2022 : 10:30 AM to 12:20 PM
• Sponsor: Survey Research Methods Section
• Author(s): Lingxiao Wang* and Yan Li and Barry Graubard and Hormuzd Katki
• Companies: National Cancer Institute, DCEG, Biostatistics Branch and University of Maryland, College Park and National Cancer Institute (NCI) and National Cancer Institute
• Keywords: risk prediction model; nonprobability cohort; finite population inference; propensity score weighting; Taylor series linearization variance

Abstract

Estimating absolute risks is fundamental to clinical decision-making but are often based on data that does not represent the target population. Current methods improve external validity by including data from population registries but require transportability assumptions of model parameters from epidemiologic studies to the population. We propose a two-step weighting procedure to estimate absolute risk in the target population without transportability assumptions. The first step improves external-validity for the cohort by creating pseudoweights for the cohort using a scaled propensity-based kernel-weighting method, which fractionally distributes sample weights from external probability reference survey units to cohort units, according to their kernel smoothed distance in propensity score. The second step poststratifies the pseudoweighted events in the cohort to a population disease registry. Our approach produces asymptotically unbiased absolute risks for the target population under correct specification of the propensity model. Poststratification improves efficiency and further reduces bias of risk estimates when the true propensity model is unknown.