Three‐phase generalized raking and multiple imputation estimators to address error‐prone data

• Published in: Statistics in medicine Vol. 43; no. 2; pp. 379 - 394
• Main Authors: Amorim, Gustavo, Tao, Ran, Lotspeich, Sarah, Shaw, Pamela A., Lumley, Thomas, Patel, Rena C., Shepherd, Bryan E.
• Format: Journal Article
• Language: English
• Published: Hoboken, USA John Wiley & Sons, Inc 30.01.2024; Wiley Subscription Services, Inc
• Subjects: Birth control; Data Accuracy; data audits; design‐based estimator; Electronic Health Records; electronic medical records; Female; HIV; HIV Infections; Human immunodeficiency virus; Humans; measurement error; Medical records; Medical statistics; multiple imputation; three‐phase design; Validation studies; Women; multiple imputation; measurement error; electronic medical records; data audits; design-based estimator; three-phase design
• ISSN: 0277-6715; 1097-0258; 1097-0258
• DOI: 10.1002/sim.9967

Validation studies are often used to obtain more reliable information in settings with error‐prone data. Validated data on a subsample of subjects can be used together with error‐prone data on all subjects to improve estimation. In practice, more than one round of data validation may be required, and direct application of standard approaches for combining validation data into analyses may lead to inefficient estimators since the information available from intermediate validation steps is only partially considered or even completely ignored. In this paper, we present two novel extensions of multiple imputation and generalized raking estimators that make full use of all available data. We show through simulations that incorporating information from intermediate steps can lead to substantial gains in efficiency. This work is motivated by and illustrated in a study of contraceptive effectiveness among 83 671 women living with HIV, whose data were originally extracted from electronic medical records, of whom 4732 had their charts reviewed, and a subsequent 1210 also had a telephone interview to validate key study variables.