Immortal Time Bias in Pharmacoepidemiology
2007/12/03 by S. Suissa, Samy Suissa · 1,712 citations
Mathematics · Medicine · #Advanced Causal Inference Techniques #Cohort #Cohort study #Confidence interval #Context (archaeology) #Epidemiology #Hazard ratio #Internal medicine #Liver Disease Diagnosis and Treatment #Mathematics #Medicine #Observational study #Pharmacoepidemiology #Pharmacology #Proportional hazards model #Statistical Methods in Clinical Trials #Statistics #Survival analysis
paper · pdf · doi:10.1093/aje/kwm324
published in American Journal of Epidemiology 167(4), 492-499 (Oxford University Press)
openalex publication_date 2007/12/03 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/04
Abstract
Immortal time is a span of cohort follow-up during which, because of exposure definition, the outcome under study could not occur. Bias from immortal time was first identified in the 1970s in epidemiology in the context of cohort studies of the survival benefit of heart transplantation. It recently resurfaced in pharmaco-epidemiology, with several observational studies reporting that various medications can be extremely effective at reducing morbidity and mortality. These studies, while using different cohort designs, all involved some form of immortal time and the corresponding bias. In this paper, the author describes various cohort study designs leading to this bias, quantifies its magnitude under different survival distributions, and illustrates it by using data from a cohort of lung cancer patients. The author shows that for time-based, event-based, and exposure-based cohort definitions, the bias in the rate ratio resulting from misclassified or excluded immortal time increases proportionately to the duration of immortal time. The bias is more pronounced with a decreasing hazard function for the outcome event, as illustrated with the Weibull distribution compared with a constant hazard from the exponential distribution. In conclusion, observational studies of drug benefit in which computerized databases are used must be designed and analyzed properly to avoid immortal time bias.
Cited by
- Impact of postoperative chemotherapy on survival for oesophagogastric adenocarcinoma after preoperative chemotherapy and surgery
- A Framework for Assessing the Viability of External Control for a Single‐Arm Trial
- Illustrating the structures of bias from immortal time using directed acyclic graphs
- Androgen deprivation therapy for prostate cancer and the risk of hospitalisation for community-acquired pneumonia
- COVID-19 mRNA Vaccination and 4-Year All-Cause Mortality Among Adults Aged 18 to 59 Years in France
- Quantifying the impact of immortal time bias: empirical evidence from meta-analyses
- Oral glucocorticoids and incidence of hypertension in people with chronic inflammatory diseases: a population-based cohort study
- European Society of Endocrinology Clinical Practice Guidelines on the management of adrenocortical carcinoma in adults, in collaboration with the European Network for the Study of Adrenal Tumors
- Causal thinking for decision making on Electronic Health Records: why and how
- Specifying a target trial prevents immortal time bias and other self-inflicted injuries in observational analyses
- The association of microhematuria with mesangial hypercellularity, endocapillary hypercellularity, crescent score and renal outcomes in immunoglobulin A nephropathy
- Implementation of the trial emulation approach in medical research: a scoping review
- Techniques to predict survival outcomes in nephrology: basics and traditional methods
- The Persistence of Retracted Papers on Wikipedia
- Cancer Risk After Radioactive Iodine Treatment for Hyperthyroidism: A Cohort Study
- Evaluating beneficial drug effects in a non‐interventional setting: a review of effectiveness studies based on Swedish Prescribed Drug Register data
- Persistent postoperative anaemia and 1‐year mortality: re‐examining time origin and sample selection
- Androgen Deprivation Therapy and the Risk of Anemia in Men with Prostate Cancer
- Utilization of Computable Phenotypes in Electronic Health Record Research: A Review and Case Study in Atopic Dermatitis
- Estimating state occupation and transition probabilities in non-Markov multi-state models subject to both random left-truncation and right-censoring
- Use of low‐dose quetiapine increases the risk of major adverse cardiovascular events: results from a nationwide active comparator‐controlled cohort study
- Challenges and new insights on estimating the effect of reintervention after EVAR
- Evaluation of Time-related Bias With Non-user Control
- Optimal timing of two-stage pulmonary metastasectomy for synchronous bilateral colorectal pulmonary metastases: A time-dependent survival analysis
- Effect of early versus deferred antiretroviral therapy for HIV on survival. [europepmc]
- Cardiovascular outcomes and mortality in patients using clopidogrel with proton pump inhibitors after percutaneous coronary intervention or acute coronary syndrome. [europepmc]
- A basic study design for expedited safety signal evaluation based on electronic healthcare data. [europepmc]
- Type 2 diabetes increases and metformin reduces total, colorectal, liver and pancreatic cancer incidences in Taiwanese: a representative population prospective cohort study of 800,000 individuals. [europepmc]
- Effect of adjuvant chemotherapy on survival of patients with stage III colon cancer diagnosed after age 75 years. [europepmc]
- Potential applications for biguanides in oncology. [europepmc]
- Propensity score methods for confounding control in nonexperimental research. [europepmc]
- Big data analysis of treatment patterns and outcomes among elderly acute myeloid leukemia patients in the United States. [europepmc]
- Using Big Data to Emulate a Target Trial When a Randomized Trial Is Not Available. [europepmc]
- Pioglitazone use and risk of bladder cancer: population based cohort study. [europepmc]
- Specifying a target trial prevents immortal time bias and other self-inflicted injuries in observational analyses. [europepmc]
- Systematic Review of Prognostic Factors for Return to Work in Workers with Sub Acute and Chronic Low Back Pain. [europepmc]
- Gestational diabetes and adverse perinatal outcomes from 716,152 births in France in 2012. [europepmc]
- AMSTAR 2: a critical appraisal tool for systematic reviews that include randomised or non-randomised studies of healthcare interventions, or both. [europepmc]
- "Resuscitation time bias"-A unique challenge for observational cardiac arrest research. [europepmc]
- Angiotensin converting enzyme inhibitors and risk of lung cancer: population based cohort study. [europepmc]
- Association of Initial Disease-Modifying Therapy With Later Conversion to Secondary Progressive Multiple Sclerosis. [europepmc]
- Deep Learning Predicts Lung Cancer Treatment Response from Serial Medical Imaging. [europepmc]
- Surgery Does Not Improve Survival in Patients With Isolated Severe Tricuspid Regurgitation. [europepmc]
- Genetic Interleukin 6 Signaling Deficiency Attenuates Cardiovascular Risk in Clonal Hematopoiesis. [europepmc]
- Ribavirin and Interferon Therapy for Critically Ill Patients With Middle East Respiratory Syndrome: A Multicenter Observational Study. [europepmc]
- Associations between stopping prescriptions for opioids, length of opioid treatment, and overdose or suicide deaths in US veterans: observational evaluation. [europepmc]
- Reflection on modern methods: trial emulation in the presence of immortal-time bias. Assessing the benefit of major surgery for elderly lung cancer patients using observational data. [europepmc]
- Evaluating eligibility criteria of oncology trials using real-world data and AI. [europepmc]
- Distinction of lymphoid and myeloid clonal hematopoiesis. [europepmc]
Related