The use of frailty hazard models for unrecognized heterogeneity that interacts with treatment

Considerations of efficiency and power

Y. Li, Rebecca Betensky, D. N. Louis, J. G. Cairncross

Research output: Contribution to journalReview article

Abstract

Increasingly, genetic studies of tumors of the same histologic diagnosis are elucidating subtypes that are distinct with respect to clinical endpoints such as response to treatment and survival. This raises concerns about the efficiency of using the simple log-rank test for analysis of treatment effect on survival in studies of possibly heterogeneous tumors. Furthermore, such studies, designed under the assumption of homogeneity, may be severely underpowered. We derive analytic approximations for the asymptotic relative efficiency of the simple log-rank test relative to the optimally weighted log-rank test and for the power of the simple log-rank test when applied to subjects with unobserved heterogeneity, as reflected in a continuous frailty, that may interact with treatment. Numerical studies demonstrate that the simple log-rank test may be quite inefficient if the frailty interacts with treatment. Further, there may be a substantial loss of power in the presence of the frailty with or without an interaction with treatment.

Original languageEnglish (US)
Pages (from-to)232-236
Number of pages5
JournalBiometrics
Volume58
Issue number1
DOIs
StatePublished - Jan 1 2002

Fingerprint

Log-rank Test
Frailty Model
Hazard Models
Proportional Hazards Models
Tumors
Hazards
Frailty
Neoplasms
testing
Tumor
Unobserved Heterogeneity
Asymptotic Relative Efficiency
neoplasms
Treatment Effects
endpoints
Homogeneity
Numerical Study
Distinct
Approximation
Interaction

Keywords

  • Omitted covariate

ASJC Scopus subject areas

  • Statistics and Probability
  • Biochemistry, Genetics and Molecular Biology(all)
  • Immunology and Microbiology(all)
  • Agricultural and Biological Sciences(all)
  • Applied Mathematics

Cite this

The use of frailty hazard models for unrecognized heterogeneity that interacts with treatment : Considerations of efficiency and power. / Li, Y.; Betensky, Rebecca; Louis, D. N.; Cairncross, J. G.

In: Biometrics, Vol. 58, No. 1, 01.01.2002, p. 232-236.

Research output: Contribution to journalReview article

@article{94524ee1a6514bbf8e4f1c51f0d1dfbe,
title = "The use of frailty hazard models for unrecognized heterogeneity that interacts with treatment: Considerations of efficiency and power",
abstract = "Increasingly, genetic studies of tumors of the same histologic diagnosis are elucidating subtypes that are distinct with respect to clinical endpoints such as response to treatment and survival. This raises concerns about the efficiency of using the simple log-rank test for analysis of treatment effect on survival in studies of possibly heterogeneous tumors. Furthermore, such studies, designed under the assumption of homogeneity, may be severely underpowered. We derive analytic approximations for the asymptotic relative efficiency of the simple log-rank test relative to the optimally weighted log-rank test and for the power of the simple log-rank test when applied to subjects with unobserved heterogeneity, as reflected in a continuous frailty, that may interact with treatment. Numerical studies demonstrate that the simple log-rank test may be quite inefficient if the frailty interacts with treatment. Further, there may be a substantial loss of power in the presence of the frailty with or without an interaction with treatment.",
keywords = "Omitted covariate",
author = "Y. Li and Rebecca Betensky and Louis, {D. N.} and Cairncross, {J. G.}",
year = "2002",
month = "1",
day = "1",
doi = "10.1111/j.0006-341X.2002.00232.x",
language = "English (US)",
volume = "58",
pages = "232--236",
journal = "Biometrics",
issn = "0006-341X",
publisher = "Wiley-Blackwell",
number = "1",

}

TY - JOUR

T1 - The use of frailty hazard models for unrecognized heterogeneity that interacts with treatment

T2 - Considerations of efficiency and power

AU - Li, Y.

AU - Betensky, Rebecca

AU - Louis, D. N.

AU - Cairncross, J. G.

PY - 2002/1/1

Y1 - 2002/1/1

N2 - Increasingly, genetic studies of tumors of the same histologic diagnosis are elucidating subtypes that are distinct with respect to clinical endpoints such as response to treatment and survival. This raises concerns about the efficiency of using the simple log-rank test for analysis of treatment effect on survival in studies of possibly heterogeneous tumors. Furthermore, such studies, designed under the assumption of homogeneity, may be severely underpowered. We derive analytic approximations for the asymptotic relative efficiency of the simple log-rank test relative to the optimally weighted log-rank test and for the power of the simple log-rank test when applied to subjects with unobserved heterogeneity, as reflected in a continuous frailty, that may interact with treatment. Numerical studies demonstrate that the simple log-rank test may be quite inefficient if the frailty interacts with treatment. Further, there may be a substantial loss of power in the presence of the frailty with or without an interaction with treatment.

AB - Increasingly, genetic studies of tumors of the same histologic diagnosis are elucidating subtypes that are distinct with respect to clinical endpoints such as response to treatment and survival. This raises concerns about the efficiency of using the simple log-rank test for analysis of treatment effect on survival in studies of possibly heterogeneous tumors. Furthermore, such studies, designed under the assumption of homogeneity, may be severely underpowered. We derive analytic approximations for the asymptotic relative efficiency of the simple log-rank test relative to the optimally weighted log-rank test and for the power of the simple log-rank test when applied to subjects with unobserved heterogeneity, as reflected in a continuous frailty, that may interact with treatment. Numerical studies demonstrate that the simple log-rank test may be quite inefficient if the frailty interacts with treatment. Further, there may be a substantial loss of power in the presence of the frailty with or without an interaction with treatment.

KW - Omitted covariate

UR - http://www.scopus.com/inward/record.url?scp=0036190971&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036190971&partnerID=8YFLogxK

U2 - 10.1111/j.0006-341X.2002.00232.x

DO - 10.1111/j.0006-341X.2002.00232.x

M3 - Review article

VL - 58

SP - 232

EP - 236

JO - Biometrics

JF - Biometrics

SN - 0006-341X

IS - 1

ER -