Methodological issues interactions

Shapiro, D and Oakley, M., Methodological issues in the evaluation of drug behavioral interactions in the treatment of hypertension. Psychosomatic Medicine 51(3), 269-276, 1989. 

Esmen, N. A., Hall, T. A., Phillips, M. L., and Marsh, G. M. (2007b). Chemical process based reconstruction of exposures for an epidemiological study. 1. Theoretical and methodological issues. Chem Biol Interact 166, 254—263. 

This brings us to the issue of risk assessment methodologies. Risk assessment is an interactive process that addresses (using risk assessment jargon) the Source, the Source-term, the Dispersion, the Dc, and the Impact a process in which there are many common components, including ... 

The methodology of surface electrochemistry is at present sufficiently broad to perform molecular-level research as required by the standards of modern surface science (1). While ultra-high vacuum electron, atom, and ion spectroscopies connect electrochemistry and the state-of-the-art gas-phase surface science most directly (1-11), their application is appropriate for systems which can be transferred from solution to the vacuum environment without desorption or rearrangement. That this usually occurs has been verified by several groups (see ref. 11 for the recent discussion of this issue). However, for the characterization of weakly interacting interfacial species, the vacuum methods may not be able to provide information directly relevant to the surface composition of electrodes in contact with the electrolyte phase. In such a case, in situ methods are preferred. Such techniques are also unique for the nonelectro-chemical characterization of interfacial kinetics and for the measurements of surface concentrations of reagents involved in... 

The linear conductance properties of a single site junction (SSJ) with Coulomb interactions (Anderson impurity model), have been extensively studied by means of the EOM approach in the cases related to CB [203,204] and the Kondo effect. [205] Later the same method was applied to some two-site models. [206-208,214] Multi-level systems were started to be considered only recently. [210,211] For out-of-equilibrium situations (finite applied bias), there are some methodological unclarified issues for calculating correlation functions using EOM techniques. [212-214] We have developed an EOM-based method which allows to deal with the finite-bias case in a self-consistent way. [209]... 

CDDs and the structurally related CDFs and dioxin-like PCBs are of concern to ATSDR because of the potential of these chemicals to harm health at relatively low doses. As discussed in Section 2.5, many of the toxic effects of these compounds appear to be mediated by a common mechanism, and CDDs frequently occur with CDFs in the environment. Therefore, due to the common mechanism of toxicity, total toxicity of a CDD/CDF mixture probably results from the added contribution (not necessarily linear) of both classes of chemicals. Because of this, the complex issue of appropriate methodology for quantitatively assessing health risks of CDDs and CDFs is currently being evaluated by ATSDR. Additional information on toxic interactions between CDDs and CDFs, as well as PCBs, would facilitate health risk assessment of this class of chemicals. 

Unlike the AT-particle picture, which in principle leads naturally to the exact solution of the many-particle Schrodinger equation, the single-particle picture has led to the development of a number of different approximation schemes designed to address particular issues in the physics of interacting quantum systems. A particularly pointed example of this state of affairs is the strict dichotomy that has set in between so-called single-particle theories and canonical many-body theory[31, 32]. Each of the two methodologies can claim a number of successful applications, which tends to reinforce the perceived formal gap between them. 

Gu and Wahba (1993) used a smoothing-spline approach with some similarities to the method described in this chapter, albeit in a context where random error is present. They approximated main effects and some specified two-variable interaction effects by spline functions. Their example had only three explanatory variables, so screening was not an issue. Nonetheless, their approach parallels the methodology we describe in this chapter, with a decomposition of a function into effects due to small numbers of variables, visualization of the effects, and an analysis of variance (ANOVA) decomposition of the total function variability. 

---