摘要:

A model-based approach is developed to estimate the distribution of time from seroconversion to diagnosis with acquired immunodeficiency syndrome (AIDS) as a function of selected time-dependent covariates. The approach is applied to longitudinal data collected over 4 years of follow-up from 450 men seropositive for the human immunodeficiency virus (90 AIDS cases) and 62 serocon-verters (nine AIDS cases) participating in the Chicago part of the Multicenter AIDS Cohort Study. Because of the periodic nature of monitoring, the seroconversion time is interval-censored for sero-converters and left-censored for seroprevalent cohort members; the end-point is right-censored for 413 individuals. Since serological monitoring is not continuous but only at regularly scheduled visit times, a model for the discrete hazard rate (DHR) is proposed that is a generalized linear model that relates the DHR to the covariate history through the complementary log-log link. Classification trees are used for preliminary screening of covariates to identify predictors of AIDS that should be incorporated into the DHR model. The missing seroconversion times for all men are imputed 100 times to obtain 100 completed datasets from which the parameters of the DHR are then estimated using the maximum-likelihood method. The final DHR model includes the following infection progression (marker) variables: CD4%, hemoglobin, p24 antigen, and CD4% X p24 antigen interaction. Using this DHR model, the discrete survival distribution of AIDS-free time is estimated for the given population. The jackknife procedure is used to assess the precision of the estimated survival distribution.

ISSN:1054-3406    

DOI:10.1080/10543409408835078

出版商:Marcel Dekker, Inc.    

年代:1994

数据来源: Taylor

摘要:

The continual reassessment method (1) (CRM) for phase I cancer trials provides improved estimation of the maximum tolerated dose (MTD), and fewer patients receive ineffective dose levels compared to the traditionally used design. However, the CRM has not gained acceptance in practice owing to concerns with administering dose levels that are too toxic. In this article, several conservative modifications of the CRM are introduced. The result is a procedure that improves estimation of the MTD and decreases the use of ineffective doses, without significantly increasing the use of toxic dose levels. The CRM with modification outperforms the traditional method in a simulation study.

ISSN:1054-3406    

DOI:10.1080/10543409408835079

出版商:Marcel Dekker, Inc.    

年代:1994

数据来源: Taylor

摘要:

Suppose, in a clinical trial, the interest is to show that the effectiveness of an experimental therapy is no worse than that of the standard therapy by more than a specified amount, say δ units. Blackwelder (1) discussed this problem in clinical trials where the outcome of interest is dichotomous. The group sequential procedures developed by DeMets and Ware (2) are valid to test Black-welder's hypothesis for the case δ = 0. In this paper, the asymmetric procedure of DeMets and Ware is modified to handle the case δ > 0. A two-stage procedure is considered in a drug interaction study that focuses on a specific side effect as the event of interest.

ISSN:1054-3406    

DOI:10.1080/10543409408835080

出版商:Marcel Dekker, Inc.    

年代:1994

数据来源: Taylor

摘要:

The statistical analysis of the repeated measures design with two factors within and no factors between subjects, which is popular in clinical pharmacology, is discussed. Use of restricted maximum likelihood (REML) methodology is compared to an imputation procedure in small sample situations with missing data and is illustrated by simulations. While imputation leads to undesirable results that are not easily corrected, REML estimation in which test statistics are compared to an F-distribution provides an elegant tool for the analysis of these designs.

ISSN:1054-3406    

DOI:10.1080/10543409408835081

出版商:Marcel Dekker, Inc.    

年代:1994

数据来源: Taylor

摘要:

The analysis of binomial data that exhibit overdispersion in a randomized complete block design is investigated. More specifically, the asymptotic distribution of the test statistic for the treatment effects in the presence of overdispersion and random effects is considered. A Monte Carlo simulation attempts to quantify the error in the approximation by theF-distribution to the distribution of the test statistic given by many popular statistics packages including SAS. The influence of different parameter values (i.e., number of treatments, number of observations per cell, degree of overdispersion, etc.) on the goodness of the approximations is evaluated. General recommendations are given for practical analysis of binomial data.

ISSN:1054-3406    

DOI:10.1080/10543409408835082

出版商:Marcel Dekker, Inc.    

年代:1994

数据来源: Taylor

摘要:

Datasets representing randomized, parallel-groups designs were analyzed by repeated measurements ANOVA and linear trend analysis with and without baseline values being covaried. ANOVA tests for the between-groups main effect, groups × times interaction, and differences in linear trends across time periods are shown to be seriously conservative or seriously nonconservative, depending on the direction and significance of chance baseline mean difference. Inclusion of baseline scores as a covariate in the repeated measurements ANOVA provides appropriate correction for the between-groups (average) effect across time, but the covariate provides no correction for the within-subject effects that are concerned with differences in the rates or patterns of change across time. If one desires to evaluate differences between patterns of treatment-induced change, tests of significance for differences in group means on composite trend scores with covariance correction for baseline are recommended. If covariance correction is not or cannot be employed, the potentially “favorable” or “unfavorable” influence of chance baseline differences on tests of significance needs to be explicitly recognized.

ISSN:1054-3406    

DOI:10.1080/10543409408835083

出版商:Marcel Dekker, Inc.    

年代:1994

数据来源: Taylor

摘要:

An analytic solution proposed by Senn (1) for removing the effects of covariate imbalance in controlled clinical trials was subjected to Monte Carlo evaluation. For practical applications of his derivation, Senn proposed substitution of sample statistics for parameters of the bivariate normal model. Unfortunately, that substitution produces severe distortion in the size of tests of significance for treatment effects when covariate imbalance is present. Numerical-verification of proposed substitutions into analytic models is recommended as a prudent approach.

ISSN:1054-3406    

DOI:10.1080/10543409408835084

出版商:Marcel Dekker, Inc.    

年代:1994

数据来源: Taylor

摘要:

Treatment side effects and associated noncompliance have methodological implications vital to the testing of new drugs. In this paper, we quantify the impact of these factors on sample size requirements in clinical trials. In the Lipid Research Clinics Trial, side effects caused treatment group compliance (50.8%) to be lower than placebo compliance (67.3%). Cholesterol reduction among treatment noncompliers was 35.2% of the reduction among compilers. Had treatment group compliance been as high as placebo compliance, 41% fewer patients would have been required to achieve the same statistical power and an expected 31% more coronary events would have been prevented. We conclude: Because they discourage patient compliance, treatment side effects can (1) cause large sample size increases, (2) lead to underestimates of true efficacy, and (3) contribute to potentially invalid negative conclusions in clinical trials. The impact of side effects goes well beyond the complications and patient discomforts with which they are associated.

ISSN:1054-3406    

DOI:10.1080/10543409408835085

出版商:Marcel Dekker, Inc.    

年代:1994

数据来源: Taylor

摘要:

Statistical considerations are discussed for the application of alternative methods to a clinical trial involving repeated ordinal ratings and multiple dosage levels of active drugs. Analyses included summary measures traditionally employed in studies of acute pain: sum of pain intensity differences from baseline, total pain relief, and total pain half gone. Estimators and confidence intervals of relative potency are developed for univariate and multivariate situations, using weighted least squares analysis with mean response and variances from Taylor series linearizations. The estimates from these methods are compared to those from traditional methods, such as ordinary least squares regression and Fieller's method for confidence intervals, as well as those from more recent developments, such as generalized estimating equations and sample survey data regression. A double-blind, two-center, randomized clinical trial of acute pain relief comparing placebo with two analgesics, each at two dosage levels, over an 8-hr period serves as an illustrative example for these techniques and comparisons.

ISSN:1054-3406    

DOI:10.1080/10543409408835086

出版商:Marcel Dekker, Inc.    

年代:1994

数据来源: Taylor

ISSN:1054-3406    

DOI:10.1080/10543409408835077

出版商:Marcel Dekker, Inc.    

年代:1994

数据来源: Taylor