Use of the landmark method to address immortal person-time bias in comparative effectiveness research: a simulation study

Xiaojuan Mi, Bradley G. Hammill, Lesley H. Curtis, Edward Chia Cheng Lai, Soko Setoguchi Iwata

Research output: Contribution to journalArticle

19 Citations (Scopus)

Abstract

Observational comparative effectiveness and safety studies are often subject to immortal person-time, a period of follow-up during which outcomes cannot occur because of the treatment definition. Common approaches, like excluding immortal time from the analysis or naïvely including immortal time in the analysis, are known to result in biased estimates of treatment effect. Other approaches, such as the Mantel–Byar and landmark methods, have been proposed to handle immortal time. Little is known about the performance of the landmark method in different scenarios. We conducted extensive Monte Carlo simulations to assess the performance of the landmark method compared with other methods in settings that reflect realistic scenarios. We considered four landmark times for the landmark method. We found that the Mantel–Byar method provided unbiased estimates in all scenarios, whereas the exclusion and naïve methods resulted in substantial bias when the hazard of the event was constant or decreased over time. The landmark method performed well in correcting immortal person-time bias in all scenarios when the treatment effect was small, and provided unbiased estimates when there was no treatment effect. The bias associated with the landmark method tended to be small when the treatment rate was higher in the early follow-up period than it was later. These findings were confirmed in a case study of chronic obstructive pulmonary disease.

Original languageEnglish (US)
Pages (from-to)4824-4836
Number of pages13
JournalStatistics in Medicine
Volume35
Issue number26
DOIs
StatePublished - Nov 20 2016

Fingerprint

Comparative Effectiveness Research
Landmarks
Person
Simulation Study
Treatment Effects
Scenarios
Estimate
Hazard
Chronic Obstructive Pulmonary Disease
Biased
Monte Carlo Simulation
Safety

All Science Journal Classification (ASJC) codes

  • Epidemiology
  • Statistics and Probability

Keywords

  • Monte Carlo simulations
  • immortal time bias
  • landmark methods
  • observational study
  • survival analysis
  • time-varying treatment

Cite this

Mi, Xiaojuan ; Hammill, Bradley G. ; Curtis, Lesley H. ; Lai, Edward Chia Cheng ; Setoguchi Iwata, Soko. / Use of the landmark method to address immortal person-time bias in comparative effectiveness research : a simulation study. In: Statistics in Medicine. 2016 ; Vol. 35, No. 26. pp. 4824-4836.
@article{6982866fc0c047a6a188b76d93409fd3,
title = "Use of the landmark method to address immortal person-time bias in comparative effectiveness research: a simulation study",
abstract = "Observational comparative effectiveness and safety studies are often subject to immortal person-time, a period of follow-up during which outcomes cannot occur because of the treatment definition. Common approaches, like excluding immortal time from the analysis or na{\"i}vely including immortal time in the analysis, are known to result in biased estimates of treatment effect. Other approaches, such as the Mantel–Byar and landmark methods, have been proposed to handle immortal time. Little is known about the performance of the landmark method in different scenarios. We conducted extensive Monte Carlo simulations to assess the performance of the landmark method compared with other methods in settings that reflect realistic scenarios. We considered four landmark times for the landmark method. We found that the Mantel–Byar method provided unbiased estimates in all scenarios, whereas the exclusion and na{\"i}ve methods resulted in substantial bias when the hazard of the event was constant or decreased over time. The landmark method performed well in correcting immortal person-time bias in all scenarios when the treatment effect was small, and provided unbiased estimates when there was no treatment effect. The bias associated with the landmark method tended to be small when the treatment rate was higher in the early follow-up period than it was later. These findings were confirmed in a case study of chronic obstructive pulmonary disease.",
keywords = "Monte Carlo simulations, immortal time bias, landmark methods, observational study, survival analysis, time-varying treatment",
author = "Xiaojuan Mi and Hammill, {Bradley G.} and Curtis, {Lesley H.} and Lai, {Edward Chia Cheng} and {Setoguchi Iwata}, Soko",
year = "2016",
month = "11",
day = "20",
doi = "https://doi.org/10.1002/sim.7019",
language = "English (US)",
volume = "35",
pages = "4824--4836",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "26",

}

Use of the landmark method to address immortal person-time bias in comparative effectiveness research : a simulation study. / Mi, Xiaojuan; Hammill, Bradley G.; Curtis, Lesley H.; Lai, Edward Chia Cheng; Setoguchi Iwata, Soko.

In: Statistics in Medicine, Vol. 35, No. 26, 20.11.2016, p. 4824-4836.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Use of the landmark method to address immortal person-time bias in comparative effectiveness research

T2 - a simulation study

AU - Mi, Xiaojuan

AU - Hammill, Bradley G.

AU - Curtis, Lesley H.

AU - Lai, Edward Chia Cheng

AU - Setoguchi Iwata, Soko

PY - 2016/11/20

Y1 - 2016/11/20

N2 - Observational comparative effectiveness and safety studies are often subject to immortal person-time, a period of follow-up during which outcomes cannot occur because of the treatment definition. Common approaches, like excluding immortal time from the analysis or naïvely including immortal time in the analysis, are known to result in biased estimates of treatment effect. Other approaches, such as the Mantel–Byar and landmark methods, have been proposed to handle immortal time. Little is known about the performance of the landmark method in different scenarios. We conducted extensive Monte Carlo simulations to assess the performance of the landmark method compared with other methods in settings that reflect realistic scenarios. We considered four landmark times for the landmark method. We found that the Mantel–Byar method provided unbiased estimates in all scenarios, whereas the exclusion and naïve methods resulted in substantial bias when the hazard of the event was constant or decreased over time. The landmark method performed well in correcting immortal person-time bias in all scenarios when the treatment effect was small, and provided unbiased estimates when there was no treatment effect. The bias associated with the landmark method tended to be small when the treatment rate was higher in the early follow-up period than it was later. These findings were confirmed in a case study of chronic obstructive pulmonary disease.

AB - Observational comparative effectiveness and safety studies are often subject to immortal person-time, a period of follow-up during which outcomes cannot occur because of the treatment definition. Common approaches, like excluding immortal time from the analysis or naïvely including immortal time in the analysis, are known to result in biased estimates of treatment effect. Other approaches, such as the Mantel–Byar and landmark methods, have been proposed to handle immortal time. Little is known about the performance of the landmark method in different scenarios. We conducted extensive Monte Carlo simulations to assess the performance of the landmark method compared with other methods in settings that reflect realistic scenarios. We considered four landmark times for the landmark method. We found that the Mantel–Byar method provided unbiased estimates in all scenarios, whereas the exclusion and naïve methods resulted in substantial bias when the hazard of the event was constant or decreased over time. The landmark method performed well in correcting immortal person-time bias in all scenarios when the treatment effect was small, and provided unbiased estimates when there was no treatment effect. The bias associated with the landmark method tended to be small when the treatment rate was higher in the early follow-up period than it was later. These findings were confirmed in a case study of chronic obstructive pulmonary disease.

KW - Monte Carlo simulations

KW - immortal time bias

KW - landmark methods

KW - observational study

KW - survival analysis

KW - time-varying treatment

UR - http://www.scopus.com/inward/record.url?scp=84976870093&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84976870093&partnerID=8YFLogxK

U2 - https://doi.org/10.1002/sim.7019

DO - https://doi.org/10.1002/sim.7019

M3 - Article

C2 - 27350312

VL - 35

SP - 4824

EP - 4836

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 26

ER -