A Comparative Analysis of Graphical Interaction and Logistic Regression Modelling: Self‐care and Coping with a Chronic Illness in Later Life

Quantitative research especially in the social, but also in the biological sciences has been limited by the availability and applicability of analytic techniques that elaborate interactions among behaviours, treatment effects, and mediating variables. This gap has been filled by a newly developed statistical technique, known as graphical interaction modelling. The merit of graphical models for analyzing highly structured data is explored in this paper by an empirical study on coping with a chronic condition as a function of interrelationships between three sets of factors. These include background factors, illness context factors, and four self‐care practices. Based on a graphical chain model, the direct and indirect dependencies are revealed and discussed in comparison to the results obtained from a simple logistic regression model ignoring possible interaction effects. Both techniques are introduced from a more tutorial point of view instead of going far into technical details.     

Immobilization of yeast cells by entrapment and adhesion using siliceous materials

Yeast cells (Saccharomyces cerevisiae) have been immobilized by entrapment in silica hydrogel, without significantly changing their biological activity; a simple model describes the rate of oxygen uptake by a film of immobilized cells. The cells have also been immobilized by direct adhesion to a glass surface; this is achieved by a well-controlled drying procedure, sufficient to bring the cells into close contact with the support, but without cell dehydration. The immobilized cells consume glucose at a rate which is about half of the rate obtained in suspension and they are resistant to strong mechanical strains.     

Time-dependent mediators in survival analysis: Modeling direct and indirect effects with the additive hazards model

We discuss causal mediation analyses for survival data and propose a new approach based on the additive hazards model. The emphasis is on a dynamic point of view, that is, understanding how the direct and indirect effects develop over time. Hence, importantly, we allow for a time varying mediator. To define direct and indirect effects in such a longitudinal survival setting we take an interventional approach (Didelez, 2018) where treatment is separated into one aspect affecting the mediator and a different aspect affecting survival. In general, this leads to a version of the nonparametric g-formula (Robins, 1986). In the present paper, we demonstrate that combining the g-formula with the additive hazards model and a sequential linear model for the mediator process results in simple and interpretable expressions for direct and indirect effects in terms of relative survival as well as cumulative hazards. Our results generalize and formalize the method of dynamic path analysis (Fosen, Ferkingstad, Borgan, & Aalen, 2006; Strohmaier et al., 2015). An application to data from a clinical trial on blood pressure medication is given.     

Covariate selection strategies for causal inference: Classification and comparison

When causal effects are to be estimated from observational data, we have to adjust for confounding. A central aim of covariate selection for causal inference is therefore to determine a set that is sufficient for confounding adjustment, but other aims such as efficiency or robustness can be important as well. In this paper, we review six general approaches to covariate selection that differ in the targeted type of adjustment set. We discuss and illustrate their advantages and disadvantages using causal diagrams. Moreover, the approaches and different ways of implementing them are compared empirically in an extensive simulation study. We conclude that there are considerable differences between the approaches but none of them is uniformly best, with performance depending on the chosen adjustment method as well as the true confounding structure. Any prior structural knowledge on the causal relations is helpful to choose the most appropriate method.