Secondary Logo

Journal Logo

Institutional members access full text with Ovid®

Censoring for loss to follow-up in time-to-event analyses of composite outcomes or in the presence of competing risks

Leskoa, Catherine R.a; Edwards, Jessie K.b; Moore, Richard D.a,c; Lau, Bryana

doi: 10.1097/EDE.0000000000001073
Original Article: PDF Only
Buy
PAP

Background In time-to-event analyses, there is limited guidance on when persons who are lost to follow-up (LTFU) should be censored.

Methods. We simulated bias in risk estimates for: (1) a composite event of measured (outcome only observable in a patient encounter) and captured events (outcome observable outside a patient encounter); and a (2) measured or (3) captured event in the presence of a competing event of the other type, under three censoring strategies: (i) censor at the last study encounter; (ii) censor when LTFU definition is met; and (iii) a new, hybrid censoring strategy. We demonstrate the real-world impact of this decision by estimating: (1) time to Acquired Immune Deficiency Syndrome diagnosis (AIDS) or death, (2) time to initiation of antiretroviral therapy (ART), and (3) time to death prior to ART initiation among adults engaged in HIV care.

Results. For (1) our hybrid censoring strategy was least biased. In our example, 5-year risk of AIDS or death was over-estimated using last-encounter censoring (25%) and under-estimated using LTFU-definition censoring (21%), compared to results from our hybrid approach (24%). Last-encounter censoring was least biased for (2). When estimating 5-year risk of ART initiation, LTFU-definition censoring underestimated risk (80% versus 85% using last-encounter censoring). LTFU-definition censoring was least biased for (3). When estimating 5-year risk of death before ART initiation, last-encounter censoring overestimated risk (5.2% versus 4.7% using LTFU-definition censoring).

Conclusions. The least biased censoring strategy for time-to-event analyses in the presence of LTFU depends on the event and estimand of interest.

a Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD

b Department of Epidemiology, University of North Carolina, Chapel Hill, NC

c Department of Medicine, Johns Hopkins School of Medicine, Baltimore, MD

Conflicts of interest: The authors have no conflicts of interest to disclose.

Simulation code is available in a supplementary web appendix.

Financial support: CRL, RDM, and BL were supported in part by National Institutes of Health grants U01 DA036935, U01 HL121812, U01 AA020793 and P30 AI094189. JKE was supported in part by NIH grants R01 AI100654.

Corresponding Author: Catherine R. Lesko, Johns Hopkins Bloomberg School of Public Health, 615 N. Wolfe St. #E7139, Baltimore, MD 21205, Phone: 410-614-6517, Email: clesko2@jhu.edu

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.