The impact of heterogeneity on the comparison of survival times
Statistics in MedicinePublished 1 October 1987
Martin Schumacher, Manfred Olschewski, Claudia Schmoor
Citations80
SJR quartileQ1
SJR score1.27
SNIP1.33
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
It is shown that heterogeneity can seriously affect the treatment comparison and has to be considered during the planning stage as well as at the analysis of a clinical trial.
Abstract
We consider several sources of heterogeneity in a clinical trial with patients' survival time as the main response criterion: differences in prognosis which can be attributed to a latent or ignored prognostic factor; differences in treatment efficacy in subgroups of patients, and differences in treatment combinations received by the patients. The impact of these types of heterogeneity on the treatment comparison is studied assuming a proportional hazards model. It is measured by the size and power of the logrank and proportional hazards score tests and by the bias of the estimated treatment effect. We show that heterogeneity can seriously affect the treatment comparison and has to be considered during the planning stage as well as at the analysis of a clinical trial.
Keywords
MathematicsEconomics, Econometrics and Finance
Journal of the Royal Statistical Society Series B (Statistical Methodology)Regression Models and Life-Tables
39,259 Citations1972D. R. Cox
BiometricsSample-Size Formula for the Proportional-Hazards Regression Model
1,029 Citations1983David Schoenfeld
A formula is derived for determining the number of observations necessary to test the equality of two survival distributions when concomitant information is incorporated, and this formula should be useful in designing clinical trials with a heterogeneous patient population.
BiometrikaSurvival models for heterogeneous populations derived from stable distributions
698 Citations1986Philip Hougaard
BiometrikaBiased estimates of treatment effect in randomized experiments with nonlinear regressions and omitted covariates
687 Citations1984Mitchell H. Gail, Sam Wieand +1 more
BiometricsTesting for Qualitative Interactions between Treatment Effects and Patient Subsets
646 Citations1985Mitchell H. Gail, Richard Simon
A likelihood ratio test is developed to test for qualitative interactions and a non-crossover interaction arises when there is variation in the magnitude, but not in the direction, of treatment effects among subsets.
BiometrikaThe asymptotic properties of nonparametric tests for comparing survival distributions
471 Citations1981David Schoenfeld
BiometrikaMisspecified proportional hazard models
380 Citations1986C. A. Struthers, John D. Kalbfleisch
BiometricsProperties of Proportional-Hazards Score Tests under Misspecified Regression Models
184 Citations1984Stephen W. Lagakos, David Schoenfeld
The effects are investigated of misspecification of the functional form of the regression portion of a proportional-hazards model on the associated partial-likelihood score test for comparing two randomized treatments in the presence of covariates.
BiometrikaA simple test of the proportional hazards assumption
177 Citations1987Richard D. Gill, Martin Schumacher
A new test of the proportional hazards assumption for two-sample censored data is presented, based on a comparison of different generalized rank estimators of the relative risk, which highlights the strengths and weaknesses of the test procedure.
British Journal of Clinical PharmacologyPatient subsets and variation in therapeutic efficacy.
132 Citations1982Richard Simon
In order to obtain the degree of reliability usually demanded of clinical therapeutic evaluations, "statistically significant' interactions between relative treatment efficacy and subsets should be demonstrated.
Statistics in MedicineThe 2 × 2 factorial design: Its application to a randomized trial of aspirin and U.S. physicians
121 Citations1985Meir J. Stampfer, Julie E. Buring +4 more
The 2 x 2 factorial design calls for randomizing each participant to treatment A or B to address one question and further assignment at random within each group to treatment C or D to examine a second issue, permitting the simultaneous test of two different hypotheses.
PubMedAlteration of cellular ribonucleases associated with murine oncogenic virus infection.
94 Citations1978Drake Wp, Frazier Lm +2 more
It is concluded that elevation of RNase activity within the lymphocyte represents an early event in oncogenic viral infection within these two tumor models.
Statistics in MedicineAssessing apparent treatment—covariate interactions in randomized clinical trials
93 Citations1985David P. Byar
The nature and interpretation of various kinds of interactions are reviewed, the use of overall tests for interaction to subset analysis is compared, and some examples of apparent treatment-covariate interactions that have arisen in actual randomized clinical trials are presented.
Journal of Statistical Computation and SimulationSample size and power determination for stratified clinical trials
89 Citations1978David P. Bernstein, Stephen W. Lagakos
An ISO FORTRAN subroutine for determining the required sample size or power in a stratified clinical trial is presented, which permits a post-accrual follow up period, censored observations, patient strata or risk groups, and unequal case allocation schemes.
BiometrikaEffect of misspecification of regression models in the analysis of survival data
88 Citations1984P. J. Solomon
Controlled Clinical TrialsInteraction between prognostic factors and treatment
52 Citations1983Jonathan J. Shuster, Jan van Eys
This article presents methodology for subdividing a target population into three subsets: one region of superiority of each treatment, and one regions of uncertainty, on the basis of treatment by prognostic factor interaction.
JAMACancer Clinical Trials: Methods and Practice
45 Citations1985Emil J. Freireich
Although "the book is aimed at clinicians and statisticians alike," the orientation is that of persons who have worked in a multi-institutional, cooperative effort, since the perspective of a major cancer center or an academic university center is not comprehensively included.
BiometricsOptimal Designs for the Analysis of Interactive Effects of Two Carcinogens or Other Toxicants
41 Citations1981Jürgen Wahrendorf, Reinhard Zentgraf +1 more
Design of animal experiments conducted to test for interaction between two carcinogens or other toxicants are considered and unbalanced designs with more animals in the combination group than in the control group are shown to provide a gain in efficacy.
Journal of Chronic DiseasesLong-term effects of interventions to improve survival in mixed populations
38 Citations1980Donald S. Shepard, Richard Zeckhauser
A general methodology is developed here for inferring the structure of a mixed population to the extent possible, for predicting accurately its response to an intervention and for extending existing models of mortality.
Statistics in MedicineAssessment of stratum‐covariate interactions in Cox's proportional hazards regression model
31 Citations1986Peter F. Thall, John M. Lachin
Striatum (e.g. treatment) interactions with covariates and with the baseline hazard function in the proportional hazards (PH) regression model for lifetime data are assessed and step-down methods for building reduced models which include stratum-specific parameters corresponding to covariates which interact with treatment are presented.
Controlled Clinical TrialsInteraction between prognostic factors and treatment
22 Citations1983Jonathan J. Shuster, Jan van Eys
CancerPerspectives on the statistician's role in cooperative clinical research
19 Citations1978Norman E. Breslow
During the final stages the major statistical contribution is in analysis and presentation of results in a way which explores the interrelationships between the most important baseline and therapeutic factors and their joint effect on outcome.
BiometricsAsymptotically Efficient Two-Step Estimators of the Hazards Ratio for Follow-Up Studies and Survival Data
16 Citations1985James R. Anderson, Leslie Bernstein
Asymptotically efficient estimators of a common hazard rate ratio (for follow-up studies) and the proportional hazards ratio ( for survival studies) are obtained by a single iteration of the "Mantel-Haenszel" estimator appropriate for each setting.
BiometrikaInteractive effects of mixtures of stimuli in life table analysis
14 Citations1983Edward L. Korn, Ping‐Yu Liu
BiometricsAssessment of Interaction between Carcinogens in Long-Term Factorially Designed Animal Experiments
5 Citations1985Stella G. Machado, Kent R. Bailey
Using Finney's underlying model, tests for interaction between carcinogens or toxic substances for data from long-term and factorially designed animal experiments are examined, based on the proportional hazards model.
Biometrical JournalModels of Independent Action for Combination Therapy
3 Citations1985Edward L. Korn
Two methods of applying Finney's concept of simple independent action are suggested for examining possible synergism in combination therapy in order to prevent the disease from progressing to a terminal outcome.