Home > Research > Publications & Outputs > Accounting for bias due to a non-ignorable trac...

Research

Associated organisational unit

Links

Text available via DOI:

Keywords

View graph of relations

Accounting for bias due to a non-ignorable tracing mechanism in a retrospective breast cancer cohort study.

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Published

Standard

Accounting for bias due to a non-ignorable tracing mechanism in a retrospective breast cancer cohort study. / Titman, Andrew C.; Lancaster, Gillian A.; Carmichael, Katie et al.
In: Statistics in Medicine, Vol. 30, No. 4, 20.02.2011, p. 324-334.

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Harvard

Titman, AC, Lancaster, GA, Carmichael, K & Scutt, D 2011, 'Accounting for bias due to a non-ignorable tracing mechanism in a retrospective breast cancer cohort study.', Statistics in Medicine, vol. 30, no. 4, pp. 324-334. https://doi.org/10.1002/sim.4118

APA

Titman, A. C., Lancaster, G. A., Carmichael, K., & Scutt, D. (2011). Accounting for bias due to a non-ignorable tracing mechanism in a retrospective breast cancer cohort study. Statistics in Medicine, 30(4), 324-334. https://doi.org/10.1002/sim.4118

Vancouver

Titman AC, Lancaster GA, Carmichael K, Scutt D. Accounting for bias due to a non-ignorable tracing mechanism in a retrospective breast cancer cohort study. Statistics in Medicine. 2011 Feb 20;30(4):324-334. doi: 10.1002/sim.4118

Author

Titman, Andrew C. ; Lancaster, Gillian A. ; Carmichael, Katie et al. / Accounting for bias due to a non-ignorable tracing mechanism in a retrospective breast cancer cohort study. In: Statistics in Medicine. 2011 ; Vol. 30, No. 4. pp. 324-334.

Bibtex

@article{0de6405565364bda893797fe173ae5aa,
title = "Accounting for bias due to a non-ignorable tracing mechanism in a retrospective breast cancer cohort study.",
abstract = "We consider the analysis of competing risks in a retrospective breast cancer cohort study where tracing of patients is dependent on survival to a pre-specified truncation time. We demonstrate that if ignored the observed cause-specific hazards will become distorted before the truncation time. Two approaches to account for the tracing bias are considered. Firstly, a likelihood based method using piecewise constant transition intensities under a Markov assumption. Secondly, a pseudo-likelihood method using inverse probability of tracing weights. For the breast cancer example, both methods improve the precision of estimates compared to a conventional approach based on excluding patients.",
keywords = "multi-state model, tracing bias, breast cancer, pseudo-likelihood",
author = "Titman, {Andrew C.} and Lancaster, {Gillian A.} and Katie Carmichael and Diane Scutt",
year = "2011",
month = feb,
day = "20",
doi = "10.1002/sim.4118",
language = "English",
volume = "30",
pages = "324--334",
journal = "Statistics in Medicine",
issn = "1097-0258",
publisher = "John Wiley and Sons Ltd",
number = "4",

}

RIS

TY - JOUR

T1 - Accounting for bias due to a non-ignorable tracing mechanism in a retrospective breast cancer cohort study.

AU - Titman, Andrew C.

AU - Lancaster, Gillian A.

AU - Carmichael, Katie

AU - Scutt, Diane

PY - 2011/2/20

Y1 - 2011/2/20

N2 - We consider the analysis of competing risks in a retrospective breast cancer cohort study where tracing of patients is dependent on survival to a pre-specified truncation time. We demonstrate that if ignored the observed cause-specific hazards will become distorted before the truncation time. Two approaches to account for the tracing bias are considered. Firstly, a likelihood based method using piecewise constant transition intensities under a Markov assumption. Secondly, a pseudo-likelihood method using inverse probability of tracing weights. For the breast cancer example, both methods improve the precision of estimates compared to a conventional approach based on excluding patients.

AB - We consider the analysis of competing risks in a retrospective breast cancer cohort study where tracing of patients is dependent on survival to a pre-specified truncation time. We demonstrate that if ignored the observed cause-specific hazards will become distorted before the truncation time. Two approaches to account for the tracing bias are considered. Firstly, a likelihood based method using piecewise constant transition intensities under a Markov assumption. Secondly, a pseudo-likelihood method using inverse probability of tracing weights. For the breast cancer example, both methods improve the precision of estimates compared to a conventional approach based on excluding patients.

KW - multi-state model

KW - tracing bias

KW - breast cancer

KW - pseudo-likelihood

U2 - 10.1002/sim.4118

DO - 10.1002/sim.4118

M3 - Journal article

VL - 30

SP - 324

EP - 334

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 1097-0258

IS - 4

ER -