Marginal and association regression models for longitudinal binary data with drop‐outs: A likelihood‐based approach

Longitudinal data often contain missing observations, and it is in general difficult to justify particular missing data mechanisms, whether random or not, that may be hard to distinguish. The authors describe a likelihood‐based approach to estimating both the mean response and association parameters for longitudinal binary data with drop‐outs. They specify marginal and dependence structures as regression models which link the responses to the covariates. They illustrate their approach using a data set from the Waterloo Smoking Prevention Project They also report the results of simulation studies carried out to assess the performance of their technique under various circumstances.     

Causal inference with generalized structural mean models

Summary.  We estimate cause–effect relationships in empirical research where exposures are not completely controlled, as in observational studies or with patient non-compliance and self-selected treatment switches in randomized clinical trials. Additive and multiplicative structural mean models have proved useful for this but suffer from the classical limitations of linear and log-linear models when accommodating binary data. We propose the generalized structural mean model to overcome these limitations. This is a semiparametric two-stage model which extends the structural mean model to handle non-linear average exposure effects. The first-stage structural model describes the causal effect of received exposure by contrasting the means of observed and potential exposure-free outcomes in exposed subsets of the population. For identification of the structural parameters, a second stage 'nuisance' model is introduced. This takes the form of a classical association model for expected outcomes given observed exposure. Under the model, we derive estimating equations which yield consistent, asymptotically normal and efficient estimators of the structural effects. We examine their robustness to model misspecification and construct robust estimators in the absence of any exposure effect. The double-logistic structural mean model is developed in more detail to estimate the effect of observed exposure on the success of treatment in a randomized controlled blood pressure reduction trial with self-selected non-compliance.     

Boundary Spaces

While shows like The X-Files and 24 have merged conspiracy theories with popular science (fictions), some video games have been pushing the narrative even further. Electronic Art's Majestic game was released in July 2001 and quickly generated media buzz with its unusual multi-modal gameplay. Mixing phone calls, faxes, instant messaging, real and "fake' websites, and email, the game provides a fascinating case of an attempt at new directions for gaming communities. Through story, mode of playing, and use of technology, Majestic highlights the uncertain status of knowledge, community and self in a digital age; at the same time, it allows examination of alternative ways of understanding games' role and purpose in the larger culture. Drawing on intricate storylines involving government conspiracies, techno-bio warfare, murder and global terror, players were asked to solve mysteries in the hopes of preventing a devastating future of domination. Because the game drew in both actual and Majestic-owned/-designed websites, it constantly pushed those playing the game right to borders where simulation collides with " factuality'. Given the wide variety of "legitimate' conspiracy theory, alien encounters and alternative science web pages, users often could not distinguish when they were leaving the game's pages and venturing into " real' World Wide Web sites. Its further use of AOL's instant messenger system, in which gamers spoke not only to bots but to other players, pushed users to evaluate constantly both the status of those they were talking to and the information being provided. Additionally, the game required players to occupy unfamiliar subject positions, ones where agency was attenuated, and which subsequently generated a multi-layered sense of unease among players. This mix of authentic and staged information in conjunction with technologically mediated roles highlights what are often seen as phenomenon endemic to the Internet itself; that is, the destabilization of categories of knowing, relating, and being.     

Measuring Substance Abuse Treatment Need Among Adults in Florida: A Social Indicators Approach

Determining the size and demographiccharacteristics of substance abuse populationsis extremely important for implementing publicpolicies aimed at the control of substanceabuse. Such information not only assists in theallocation of limited treatment resources bythe state, but also in the monitoring ofsubstance abuse trends over time and in theevaluation of innovative policy initiatives. Inthis study, we develop three composite measuresof treatment need. We then use these measuresto estimate treatment need for alcohol abuseand for controlled substance abuse within eachof Florida's 67 counties. This study providesan important empirical component of communityplanning, quantifying and, to a limited degree,specifying the level of need for the substanceabuse treatment of community residents. Anadditional benefit is the development of a costeffective and unobtrusive methodology fordetermining empirically when levels of need arechanging so that treatment levels can beadjusted accordingly. With proper use,policymakers can readily employ the methodologydeveloped in this study in Florida andelsewhere to make better-informed decisions inthe allocation of finite substance abusetreatment resources.     

Scheduling Updates of Probabilistic Risk Assessments: The Arkansas Nuclear One-Unit 1 Experience

This paper presents the results of a study that identified how often a probabilistic risk assessment (PRA)should be updated to accommodate the changes that take place at nuclear power plants. Based on a 7-year analysis of design and procedural changes at one plant, we consider 5 years to be the maximum interval for updating PRAs. This conclusion is preliminary because it is based on the review of changes that occurred at a single plant, and it addresses only PRAs that involve a Level 1 analysis (i.e., a PRA including calculation of core damage frequency only). Nevertheless, this conclusion indicates that maintaining a useful PRA requires periodic updating efforts. However, the need for this periodic update stems only partly from the number of changes that can be expected to take place at nuclear power plants–changes that individually have only a moderate to minor impact on the PRA, but whose combined impact is substantial and necessitates a PRA update. Additionally, a comparison of two generations of PRAs performed about 5 years apart indicates that PRAs must be periodically updated to reflect the evolution of PRA methods. The most desirable updating interval depends on these two technical considerations as well as the cost of updating the PRA. (Cost considerations, however, were beyond the scope of this study.)     

The analysis of incomplete data using stochastic covariates

In the development of many diseases there are often associated random variables which continuously reflect the progress of a subject towards the final expression of the disease (failure). At any given time these processes, which we call stochastic covariates, may provide information about the current hazard and the remaining time to failure. Likewise, in situations when the specific times of key prior events are not known, such as the time of onset of an occult tumour or the time of infection with HIV-1, it may be possible to identify a stochastic covariate which reveals, indirectly, when the event of interest occurred. The analysis of carcinogenicity trials which involve occult tumours is usually based on the time of death or sacrifice and an indicator of tumour presence for each animal in the experiment. However, the size of an occult tumour observed at the endpoint represents data concerning tumour development which may convey additional information concerning both the tumour incidence rate and the rate of death to which tumour-bearing animals are subject. We develop a stochastic model for tumour growth and suggest different ways in which the effect of this growth on the hazard of failure might be modelled. Using a combined model for tumour growth and additive competing risks of death, we show that if this tumour size information is used, assumptions concerning tumour lethality, the context of observation or multiple sacrifice times are no longer necessary in order to estimate the tumour incidence rate. Parametric estimation based on the method of maximum likelihood is outlined and is applied to simulated data from the combined model. The results of this limited study confirm that use of the stochastic covariate tumour size results in more precise estimation of the incidence rate for occult tumours.