Biomedical subjects

Anastasios A Tsiatis

Publications and source records attributed to Anastasios A Tsiatis.

16 recordsLinked to original sources

PubMed2006

Adaptive two-stage designs in phase II clinical trials.

Two-stage designs have been widely used in phase II clinical trials. Such designs are desirable because they allow a decision to be made on whether a treatment is effective or not after the accumulation of the data at the end of each stage. Optimal fixed two-stage designs, where the sample size at each stage is fixed in advance, were proposed by Simon when the primary outcome is a binary response. This paper proposes an adaptive two-stage design which allows the sample size at the second stage to depend on the results at the first stage. Using a Bayesian decision-theoretic construct, we derive optimal adaptive two-stage designs; the optimality criterion being minimum expected sample size under the null hypothesis. Comparisons are made between Simon's two-stage fixed design and the new design with respect to this optimality criterion.

Clinical Trials, Phase II as Topic↗

PubMed2006

Information-based monitoring of clinical trials.

When designing a clinical trial to compare the effect of different treatments on response, a key issue facing the statistician is to determine how large a study is necessary to detect a clinically important difference with sufficient power. This is the case whether the study will be analysed only once (single-analysis) or whether it will be monitored periodically with the possibility of early stopping (group-sequential). Standard sample size calculations are based on both the magnitude of difference that is considered clinically important as well as values for the nuisance parameters in the statistical model. For planning purposes, best guesses are made for the value of the nuisance parameters and these are used to determine the sample size. However, if these guesses are incorrect this will affect the subsequent power to detect the clinically important difference. It is argued in this paper that statistical precision is directly related to Statistical Information and that the study should continue until the requisite statistical information is obtained. This is referred to as information-based design and analysis of clinical trials. We also argue that this type of methodology is best suited with group-sequential trials which monitor the data periodically and allow for estimation of the statistical information as the study progresses.

Clinical Trials Data Monitoring Committees↗

PubMed2006

Cost-effectiveness of defibrillator therapy or amiodarone in chronic stable heart failure: results from the Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT).

BACKGROUND: In the Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT), implantable cardioverter-defibrillator (ICD) therapy significantly reduced all-cause mortality rates compared with medical therapy alone in patients with stable, moderately symptomatic heart failure, whereas amiodarone had no benefit on mortality rates. We examined long-term economic implications of these results. METHODS AND RESULTS: Medical costs were estimated by using hospital billing data and the Medicare Fee Schedule. Our base case cost-effectiveness analysis used empirical clinical and cost data to estimate the lifetime incremental cost of saving an extra life-year with ICD therapy relative to medical therapy alone. At 5 years, the amiodarone arm had a survival rate equivalent to that of the placebo arm and higher costs than the placebo arm. For ICD relative to medical therapy alone, the base case lifetime cost-effectiveness and cost-utility ratios (discounted at 3%) were dollar 38,389 per life-year saved (LYS) and dollar 41,530 per quality-adjusted LYS, respectively. A cost-effectiveness ratio < dollar 100,000 was obtained in 99% of 1000 bootstrap repetitions. The cost-effectiveness ratio was sensitive to the amount of extrapolation beyond the empirical 5-year trial data: dollar 127,503 per LYS at 5 years, dollar 88,657 per LYS at 8 years, and dollar 58,510 per LYS at 12 years. Because of a significant interaction between ICD treatment and New York Heart Association class, the cost-effectiveness ratio was dollar 29,872 per LYS for class II, whereas there was incremental cost but no incremental benefit in class III. CONCLUSIONS: Prophylactic use of single-lead, shock-only ICD therapy is economically attractive in patients with stable, moderately symptomatic heart failure with an ejection fraction < or = 35%, particularly those in NYHA class II, as long as the benefits of ICD therapy observed in the SCD-HeFT persist for at least 8 years.

Adult↗

PubMed2005

Comparison between two partial likelihood approaches for the competing risks model with missing cause of failure.

In many clinical studies where time to failure is of primary interest, patients may fail or die from one of many causes where failure time can be right censored. In some circumstances, it might also be the case that patients are known to die but the cause of death information is not available for some patients. Under the assumption that cause of death is missing at random, we compare the Goetgbebeur and Ryan (1995, Biometrika, 82, 821-833) partial likelihood approach with the Dewanji (1992, Biometrika, 79, 855-857) partial likelihood approach. We show that the estimator for the regression coefficients based on the Dewanji partial likelihood is not only consistent and asymptotically normal, but also semiparametric efficient. While the Goetghebeur and Ryan estimator is more robust than the Dewanji partial likelihood estimator against misspecification of proportional baseline hazards, the Dewanji partial likelihood estimator allows the probability of missing cause of failure to depend on covariate information without the need to model the missingness mechanism. Tests for proportional baseline hazards are also suggested and a robust variance estimator is derived.

Cause of Death↗

PubMed2005

Conducting economic evaluations alongside multinational clinical trials: toward a research consensus.

Demand for economic evaluations in multinational clinical trials is increasing, but there is little consensus about how such studies should be conducted and reported. At a workshop in Durham, North Carolina, we sought to identify areas of agreement about how the primary findings of economic evaluations in multinational clinical trials should be generated and presented. In this paper, we propose a framework for classifying multinational economic evaluations according to (a) the sources of an analyst's estimates of resource use and clinical effectiveness and (b) the analyst's method of estimating costs. We review existing studies in the cardiology literature in the context of the proposed framework. We then describe important methodological and practical considerations in conducting multinational economic evaluations and summarize the advantages and disadvantages of each approach. Finally, we describe opportunities for future research. Delineation of the various approaches to multinational economic evaluation may assist researchers, peer reviewers, journal editors, and decision makers in evaluating the strengths and limitations of particular studies.

Clinical Trials as Topic↗

PubMed2004

Optimal duration of eptifibatide infusion in percutaneous coronary intervention (an ESPRIT substudy).

Although randomized trials have clearly demonstrated the clinical efficacy with regimens of platelet glycoprotein IIb/IIIa antagonists that result in >80% inhibition of baseline platelet aggregation in percutaneous coronary intervention (PCI), there are no data available concerning the optimal duration of infusion of these agents. In an era when the length of hospitalization has a major impact on health care costs, the determination of the optimal duration of the infusion of these drugs after PCI is of great relevance. The investigators therefore sought to determine the optimal length of the infusion of eptifibatide after PCI by analyzing the outcomes of patients enrolled in the Enhanced Suppression of the Platelet IIb/IIIa Receptor With Integrilin Therapy trial who were randomized to treatment with eptifibatide.

Angioplasty, Balloon, Coronary↗

PubMed2004

Marginal structural models for analyzing causal effects of time-dependent treatments: an application in perinatal epidemiology.

Marginal structural models (MSMs) are causal models designed to adjust for time-dependent confounding in observational studies of time-varying treatments. MSMs are powerful tools for assessing causality with complicated, longitudinal data sets but have not been widely used by practitioners. The objective of this paper is to illustrate the fitting of an MSM for the causal effect of iron supplement use during pregnancy (time-varying treatment) on odds of anemia at delivery in the presence of time-dependent confounding. Data from pregnant women enrolled in the Iron Supplementation Study (Raleigh, North Carolina, 1997-1999) were used. The authors highlight complexities of MSMs and key issues epidemiologists should recognize before and while undertaking an analysis with these methods and show how such methods can be readily interpreted in existing software packages, including SAS and Stata. The authors emphasize that if a data set with rich information on confounders is available, MSMs can be used straightforwardly to make robust inferences about causal effects of time-dependent treatments/exposures in epidemiologic research.

Adult↗

PubMed2004

The prognostic importance of comorbidity for mortality in patients with stable coronary artery disease.

OBJECTIVES: To identify the prevalent and prognostically important coexisting illnesses among single coronary artery disease (CAD) patients. BACKGROUND: As the population ages, physicians are increasingly required to make decisions concerning patients with multiple co-existing illnesses (comorbidity). Many trials of CAD therapy have excluded patients with significant comorbidity, such that there are limited data to guide the management of those patients. METHODS: To consider the long-term prognostic importance of comorbid illness, we examined a cohort of 1471 patients with CAD who underwent cardiac catheterization between 1985 and 1989 and were followed up through 2000 in the Duke Databank for Cardiovascular Diseases. Weights were assigned to individual diseases according to their prognostic significance in Cox proportional hazards models, thus creating a new CAD-specific index. The new index was compared with the widely used Charlson index, according to prevalence of conditions, individual and overall associations with survival, and agreement. RESULTS: The Charlson index and the CAD-specific index were highly associated with long-term survival and almost equivalent to left ventricular ejection fraction. When considering the components of the Charlson index, diabetes, renal insufficiency, chronic obstructive pulmonary disease, and peripheral vascular disease had greater prognostic significance among CAD patients, whereas peptic ulcer disease, connective tissue disease, and lymphoma were less significant. Hemiplegia, leukemia, lymphoma, severe liver disease, and acquired immunodeficiency syndrome were rarely identified among patients undergoing coronary angiography. CONCLUSIONS: Comorbid disease is strongly associated with long-term survival in patients with CAD. These data suggest co-existing illnesses should be measured and considered in clinical trials, disease registries, quality comparisons, and counseling of individual patients.

Black People↗

PubMed2004

Differential treatment benefit of platelet glycoprotein IIb/IIIa inhibition with percutaneous coronary intervention versus medical therapy for acute coronary syndromes: exploration of methods.

BACKGROUND: Although many believe that platelet glycoprotein IIb/IIIa inhibitors should be used only in acute coronary syndrome patients undergoing percutaneous coronary intervention, supporting data from randomized clinical trials are tenuous. The assumption that these agents are useful only in conjunction with percutaneous coronary intervention is based primarily on inappropriate subgroup analyses performed across the glycoprotein IIb/IIIa inhibitor trials. METHODS AND RESULTS: We describe the problems with these analytical techniques and demonstrate that different approaches to the question can result in opposing answers. CONCLUSIONS: Clinical-practice decisions and practice guidelines should be based on overall trial results and not analyses of post-randomization subgroups.

Acute Disease↗

PubMed2004

Optimal estimator for the survival distribution and related quantities for treatment policies in two-stage randomization designs in clinical trials.

Two-stage designs, where patients are initially randomized to an induction therapy and then depending upon their response and consent, are randomized to a maintenance therapy, are common in cancer and other clinical trials. The goal is to compare different combinations of primary and maintenance therapies to find the combination that is most beneficial. In practice, the analysis is usually conducted in two separate stages which does not directly address the major objective of finding the best combination. Recently Lunceford, Davidian, and Tsiatis (2002, Biometrics58, 48-57) introduced ad hoc estimators for the survival distribution and mean restricted survival time under different treatment policies. These estimators are consistent but not efficient, and do not include information from auxiliary covariates. In this article we derive estimators that are easy to compute and are more efficient than previous estimators. We also show how to improve efficiency further by taking into account additional information from auxiliary variables. Large sample properties of these estimators are derived and comparisons with other estimators are made using simulation. We apply our estimators to a leukemia clinical trial data set that motivated this study.

Antineoplastic Agents↗

PubMed2004

Estimating mean response as a function of treatment duration in an observational study, where duration may be informatively censored.

After a treatment is found to be effective in a clinical study, attention often focuses on the effect of treatment duration on outcome. Such an analysis facilitates recommendations on the most beneficial treatment duration. In many studies, the treatment duration, within certain limits, is left to the discretion of the investigators. It is often the case that treatment must be terminated prematurely due to an adverse event, in which case a recommended treatment duration is part of a policy that treats patients for a specified length of time or until a treatment-censoring event occurs, whichever comes first. Evaluating mean response for a particular treatment-duration policy from observational data is difficult due to censoring and the fact that it may not be reasonable to assume patients are prognostically similar across all treatment strategies. We propose an estimator for mean response as a function of treatment-duration policy under these conditions. The method uses potential outcomes and embodies assumptions that allow consistent estimation of the mean response. The estimator is evaluated through simulation studies and demonstrated by application to the ESPRIT infusion trial coordinated at Duke University Medical Center.

Biometry↗

PubMed2003

Semiparametric estimation of treatment effect in a pretest-posttest study.

Inference on treatment effects in a pretest-posttest study is a routine objective in medicine, public health, and other fields. A number of approaches have been advocated. We take a semiparametric perspective, making no assumptions about the distributions of baseline and posttest responses. By representing the situation in terms of counterfactual random variables, we exploit recent developments in the literature on missing data and causal inference, to derive the class of all consistent treatment effect estimators, identify the most efficient such estimator, and outline strategies for implementation of estimators that may improve on popular methods. We demonstrate the methods and their properties via simulation and by application to a data set from an HIV clinical trial.

Antiviral Agents↗

PubMed2002

An estimator for the proportional hazards model with multiple longitudinal covariates measured with error.

In many longitudinal studies, it is of interest to characterize the relationship between a time-to-event (e.g. survival) and several time-dependent and time-independent covariates. Time-dependent covariates are generally observed intermittently and with error. For a single time-dependent covariate, a popular approach is to assume a joint longitudinal data-survival model, where the time-dependent covariate follows a linear mixed effects model and the hazard of failure depends on random effects and time-independent covariates via a proportional hazards relationship. Regression calibration and likelihood or Bayesian methods have been advocated for implementation; however, generalization to more than one time-dependent covariate may become prohibitive. For a single time-dependent covariate, Tsiatis and Davidian (2001) have proposed an approach that is easily implemented and does not require an assumption on the distribution of the random effects. This technique may be generalized to multiple, possibly correlated, time-dependent covariates, as we demonstrate. We illustrate the approach via simulation and by application to data from an HIV clinical trial.

Journal Article↗

PubMed2002

Estimation of survival distributions of treatment policies in two-stage randomization designs in clinical trials.

Some clinical trials follow a design where patients are randomized to a primary therapy at entry followed by another randomization to maintenance therapy contingent upon disease remission. Ideally, analysis would allow different treatment policies, i.e., combinations of primary and maintenance therapy if specified up-front, to be compared. Standard practice is to conduct separate analyses for the primary and follow-up treatments, which does not address this issue directly. We propose consistent estimators for the survival distribution and mean restricted survival time for each treatment policy in such two-stage studies and derive large-sample properties. The methods are demonstrated on a leukemia clinical trial data set and through simulation.

Aged↗

PubMed2002

Median regression with censored cost data.

Because of the skewness of the distribution of medical costs, we consider modeling the median as well as other quantiles when establishing regression relationships to covariates. In many applications, the medical cost data are also right censored. In this article, we propose semiparametric procedures for estimating the parameters in median regression models based on weighted estimating equations when censoring is present. Numerical studies are conducted to show that our estimators perform well with small samples and the resulting inference is reliable in circumstances of practical importance. The methods are applied to a dataset for medical costs of patients with colorectal cancer.

Biometry↗

PubMed2002

A semiparametric likelihood approach to joint modeling of longitudinal and time-to-event data.

Joint models for a time-to-event (e.g., survival) and a longitudinal response have generated considerable recent interest. The longitudinal data are assumed to follow a mixed effects model, and a proportional hazards model depending on the longitudinal random effects and other covariates is assumed for the survival endpoint. Interest may focus on inference on the longitudinal data process, which is informatively censored, or on the hazard relationship. Several methods for fitting such models have been proposed, most requiring a parametric distributional assumption (normality) on the random effects. A natural concern is sensitivity to violation of this assumption; moreover, a restrictive distributional assumption may obscure key features in the data. We investigate these issues through our proposal of a likelihood-based approach that requires only the assumption that the random effects have a smooth density. Implementation via the EM algorithm is described, and performance and the benefits for uncovering noteworthy features are illustrated by application to data from an HIV clinical trial and by simulation.

Anti-HIV Agents↗