Methods in Pharmacoepidemiology / Confounding/Bias

6F-04 - Bias Due to Interval Censored Outcomes in a Study of Flare Risk after Hydroxychloroquine Taper/Cessation in Systemic Lupus Erythematosus

Tuesday, August 26, 2025

2:15 PM - 2:30 PM ET

Presenting Author(s)

RK

Rima Kaddoura

McGill University, Canada

Co-Author(s)

SB

Sasha Bernatsky

Centre for Outcomes Research and Evaluation, The Research Institute of the McGill University Health Centre, Canada
CA

Celline C. Almeida-Brasil

Centre for Outcomes Research and Evaluation (CORE), The Research Institute of the McGill University Health Centre, Canada
JH

John G. Hanly

Rheumatology, Dalhousie University, Canada
MU

Murray Urowitz

Schroeder Arthritis Institute, Krembil Research Institute, Center for Prognosis Studies in the Rheumatic Diseases, Toronto Western Hospital, University of Toronto, Canada
AC

Ann E. Clarke

Division of Rheumatology, Cumming School of Medicine, University of Calgary, Canada
GR

Guillermo Ruiz-Irastorza

Autoimmune Diseases Research Unit, Biocruces Bizkaia Health Research Institute, Spain
CG

Caroline Gordon

Rheumatology Research Group, University of Birmingham, United Kingdom
RR

Rosalind Ramsey-Goldman

Medicine/Rheumatology, Northwestern University, United States
MP

Michelle A. Petri

Division of Rheumatology, Johns Hopkins University School of Medicine, United States
EG

Ellen M. Ginzler

Medicine/Rheumatology, SUNY Downstate Medical Center, United States
DW

Daniel J. Wallace

Cedars-Sinai/David Geffen School of Medicine at UCLA, Cedars-Sinai Medical Center, United States
SB

Sang-Cheol Bae

Rheumatology, Hanyang University Institute for Rheumatology and Hanyang University Institute of Bioscience and Biotechnology, Republic of Korea
JR

Juanita Romero-Diaz

Immunology and Rheumatology, Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubirán, Mexico
MD

Mary-Anne Dooley

Rheumatology, The University of North Carolina at Chapel Hill, United States
CP

Christine Peschken

Rheumatology, University of Manitoba, Canada
DI

David Isenberg

Medicine (Rheumatology), University College London, United Kingdom
SM

Susan Manzi

Allegheny Singer Research Institute, Allegheny Health Network, United States
SJ

Søren Jacobsen

Copenhagen Lupus and Vasculitis Clinic, Copenhagen University Hospital Center for Rheumatology and Spine Diseases, Denmark
SL

S Sam Lim

School of Medicine, Emory University, United States
ON

Ola Nived

Department of Clinical Science, Rheumatology, Lund University Faculty of Medicine, Sweden
AJ

Andreas Jönsen

Department of Clinical Science, Rheumatology, Lund University Faculty of Medicine, Sweden
DK

Diane L. Kamen

Medicine, Medical University of South Carolina, United States
CA

Cynthia Aranow

Autoimmune and Musculoskeletal Disease, Feinstein Institute for Medical Research Fertility Research Laboratory, United States
JS

Jorge Sánchez-Guerrero

Center for Prognosis Studies in the Rheumatic Diseases, Toronto Western Hospital, Canada
DG

Dafna D. Gladman

Center for Prognosis Studies in the Rheumatic Diseases, Toronto Western Hospital, Canada
PF

Paul R. Fortin

Medicine-Rheumatology, Universite Laval, Canada
GA

Graciela S. Alarcon

Medicine, Division of Clinical Immunology and Rheumatology, The University of Alabama at Birmingham, United States
JM

Joan T. Merrill

Arthritis & Clinical Immunology Program, Oklahoma Medical Research Foundation, United States
KK

Kenneth Kalunian

Division of Rheumatology, Allergy and Immunology, University of California San Diego, United States
MR

Manuel Ramos-Casals

Department of Autoimmune Diseases, Hospital Clínic, Universitat de Barcelona, Spain
AZ

Asad Zoma

Lanarkshire Centre for Rheumatology, Hairmyres Hospital, United Kingdom
AA

Anca D. Askanase

Rheumatology, Columbia University Medical Center, Canada
MK

Munther Khamashta

Rayne Institute, St Thomas Hospital, United Kingdom
IB

Ian N. Bruce

Arc Epidemiology Unit, The University of Manchester, United Kingdom
MI

Murat Inanc

Internal Medicine Rheumatology, Istanbul University, Turkey
LL

Luck Lukusa

Centre for Outcomes Research and Evaluation (CORE), Research Institute of the McGill University Health Centre, Canada
MA

Michal Abrahamowicz

Department of Epidemiology, Biostatistics and Occupational Health, Faculty of Medicine and Health Sciences, McGill University, Canada

Background: In survival analysis, interval-censoring (IC) occurs when an outcome can only be determined at set time-points, not at the actual event time. This biases hazard ratio (HR) estimates toward the null.

Objectives: We estimated bias due to IC using real-world data from the Systemic Lupus International Collaborating Clinics inception cohort using data-driven simulations.

Methods: The time-varying exposure of interest was a binary indicator of hydroxychloroquine (HCQ) tapering/cessation. The composite outcome was lupus flare, defined by annual disease scores and medication changes. Components of the outcome were IC, as they were recorded only annually, without a precise event date. For the unknown IC event times, a “true” event time was randomly generated from a uniform distribution of the time between two assessments. Each simulated sample was analyzed separately using event times (for IC outcomes) imputed either at the midpoint or the later of the two adjacent yearly assessments. Then, SIMulation EXtrapolation (SIMEX) methods were used to correct the naïve adjusted HR estimates, adjusted for demographics, drugs, and clinical variables.

Results: The 1,543 subjects were followed for a median of 26 months. During follow-up, 398 participants tapered/stopped HCQ and 1,187 had a flare. The adjusted uncorrected HR was 1.53 (95% CI 1.32-1.78) when IC event times were imputed at the mid-point and 1.43 (95% CI 1.24-1.66) when imputed at the later of the two assessments. Data-driven simulations showed that IC-related bias toward the null is systematically smaller with mid-point imputation. SIMEX correction yielded a higher point estimate for the adjusted HR (1.68, bootstrapped 95% CI 1.44-2.02).

Conclusions: IC events induced bias toward the null in our HR estimates for the association between HCQ taper/cessation and lupus flares. SIMEX correction can mitigate this bias.