Search

In survival studies, information lost through censoring can be partially recaptured through repeated measures data which are predictive of survival. In addition, such data may be useful in removing bias in survival estimates, due to censoring which depends upon the repeated measures. Here we investigate joint models for survival T and repeated measurements Y, given a vector of covariates Z. Mixture models indexed as f (T/Z) f (Y/T,Z) are well suited for assessing covariate effects on survival time. Our objective is efficiency gains, using non-parametric models for Y in order to avoid introducing bias by misspecification of the distribution for Y. We model (T/Z) as a piecewise exponential distribution with proportional hazards covariate effect. The component (Y/T,Z) has a multinomial model. The joint likelihood for survival and longitudinal data is maximized, using the EM algorithm. The estimate of covariate effect is compared to the estimate based on the standard proportional hazards model and an alternative joint model based estimate. We demonstrate modest gains in efficiency when using the joint piecewise exponential joint model. In a simulation, the estimated efficiency gain over the standard proportional hazards model is 6.4 per cent. In clinical trial data, the estimated efficiency gain over the standard proportional hazards model is 10.2 per cent., (Copyright 2003 John Wiley & Sons, Ltd.)

This paper considers the mixture model methodology for handling non-ignorable drop-outs in longitudinal studies with continuous outcomes. Recently, Hogan and Laird have developed a mixture model for non-ignorable drop-outs which is a standard linear mixed effects model except that the parameters which characterize change over time depend also upon time of drop-out. That is, the mean response is linear in time, other covariates and drop-out time, and their interactions. One of the key attractions of the mixture modelling approach to drop-outs is that it is relatively easy to explore the sensitivity of results to model specification. However, the main drawback of mixture models is that the parameters that are ordinarily of interest are not immediately available, but require marginalization of the distribution of outcome over drop-out times. Furthermore, although a linear model is assumed for the conditional mean of the outcome vector given time of drop out, after marginalization, the unconditional mean of the outcome vector is not, in general, linear in the regression parameters. As a result, it is not possible to parsimoniously describe the effects of covariates on the marginal distribution of the outcome in terms of regression coefficients. The need to explicitly average over the distribution of the drop-out times and the absence of regression coefficients that describe the effects of covariates on the outcome are two unappealing features of the mixture modelling approach. In this paper we describe a particular parameterization of the general linear mixture model that circumvents both of these problems., (Copyright 2001 John Wiley & Sons, Ltd.)

The general linear mixed model provides a useful approach for analysing a wide variety of data structures which practising statisticians often encounter. Two such data structures which can be problematic to analyse are unbalanced repeated measures data and longitudinal data. Owing to recent advances in methods and software, the mixed model analysis is now readily available to data analysts. The model is similar in many respects to ordinary multiple regression, but because it allows correlation between the observations, it requires additional work to specify models and to assess goodness-of-fit. The extra complexity involved is compensated for by the additional flexibility it provides in model fitting. The purpose of this tutorial is to provide readers with a sufficient introduction to the theory to understand the method and a more extensive discussion of model fitting and checking in order to provide guidelines for its use. We provide two detailed case studies, one a clinical trial with repeated measures and dropouts, and one an epidemiological survey with longitudinal follow-up.

Many long-term clinical trials collect both a vector of repeated measurements and an event time on each subject; often, the two outcomes are dependent. One example is the use of surrogate markers to predict disease onset or survival. Another is longitudinal trials which have outcome-related dropout. We describe a mixture model for the joint distribution which accommodates incomplete repeated measures and right-censored event times, and provide methods for full maximum likelihood estimation. The methods are illustrated through analysis of data from a clinical trial for a new schizophrenia therapy; in the trial, dropout time is closely related to outcome, and the dropout process differs between treatments. The parameter estimates from the model are used to make a treatment comparison after adjusting for the effects of dropout. An added benefit of the analysis is that it permits using the repeated measures to increase efficiency of estimates of the event time distribution.

Since Wu and Carroll (Biometrics 44, 175-188) proposed a model for longitudinal progression in the presence of informative dropout, several researchers have developed and studied models for situations where both a vector of repeated outcomes and an event time is available for each subject. These models have been developed for either longitudinal studies with dropout or for survival studies in which a random, time-varying covariate is measured repeatedly across time. When inference about the longitudinal variable is of interest, event times are treated as covariates and are often incomplete due to censoring. If survival or event time is the primary endpoint, repeated outcomes observed prior to the event are viewed as covariates; this covariate process is often incomplete, measured with error, or observed at unscheduled times during the study. We review several models which are used to handle incomplete response and covariate data in both survival and longitudinal studies.

The crossover design is a type of longitudinal study with subjects receiving different treatments in different time periods. When carry-over effects are absent, the usual crossover design is structured so that all the information about treatment effects is contained in the within-subject contrasts; standard analyses are based on these within-subject contrasts and ignore any between-subject information. With carry-over effects present these standard analyses can be very inefficient, especially for suboptimal designs. We describe alternative approaches based on methods for the analysis of longitudinal data.

We discuss maximum likelihood methods for analysing binary responses measured at two times, such as in a cross-over design. We construct a 2 x 2 table for each individual with cell probabilities corresponding to the cross-classification of the responses at the two times; the underlying likelihood for each individual is multinomial with four cells. The three dimensional parameter space of the multinomial distribution is completely specified by the two marginal probabilities of success of the 2 x 2 table and an association parameter between the binary responses at the two times. We examine a logistic model for the marginal probabilities of the 2 x 2 table for individual i; the association parameters we consider are either the correlation coefficient, the odds ratio or the relative risk. Simulations show that the parameter estimates for the logistic regression model for the marginal probabilities are not very sensitive to the parameters used to describe the association between the binary responses at the two times. Thus, we suggest choosing the measure of association for ease of interpretation.

When observations are made repeatedly over time on the same experimental units, unbalanced patterns of observations are a common occurrence. This complication makes standard analyses more difficult or inappropriate to implement, means loss of efficiency, and may introduce bias into the results as well. Some possible approaches to dealing with missing data include complete case analyses, univariate analyses with adjustments for variance estimates, two-step analyses, and likelihood based approaches. Likelihood approaches can be further categorized as to whether or not an explicit model is introduced for the non-response mechanism. This paper will review the use of likelihood based analyses for longitudinal data with missing responses, both from the point of view of ease of implementation and appropriateness in view of the non-response mechanism. Models for both measured and dichotomous outcome data will be discussed. The appropriateness of some non-likelihood based analyses is briefly considered.

We compare two statistical methods for combining event rates from several studies. Both methods treat each study as a separate stratum. The Peto-modified Mantel-Haenszel (Peto) method estimates a combined odds ratio assuming homogeneity across strata and provides a test for heterogeneity. The DerSimonian and Laird modified Cochran method (D&L) produces a weighted average of rate differences, where the weights allow for among-study variability. We analyse 22 meta-analyses from ten reports by both methods. The pooled estimates are divided by their standard errors to produce a Z-statistic. A t-test comparing Z-statistics from all 22 studies suggests that the D&L method tends to be more conservative [d(Peto - D&L) = 0.29, t = 2.53, p = 0.02]. For a subset of 14 non-heterogeneous studies, the difference is smaller and non-significant (d = 0.09, t = 0.72, p = 0.49). The results from the methods correlate well (r = 0.66 for all 22 studies, r = 0.95 for 14 non-heterogeneous studies). Thus, the presence of heterogeneity influences our conclusion. We discuss the statistical and scientific implications of these findings.