Risk analysis. Foundations, models, and methods. (Q1422058): Difference between revisions

Health risk analysis comprises methods, procedures and principles for answering (for example) questions of the following type: 1) Does a particular chemical substitute cause cancer or some other disease? 2) How does the risk of a disease depend on the concentration and the time of exposure to a potential risk factor? 3) How much compensation should a company pay to an employee who develops cancer if the employee was exposed to radiation but also had non-occupational risk factors, such as smoking, poor diet and medical radiation? 4) What are the best countermeasures for the problem of antibiotic-resistant strains of bacteria which may survive in undercooked meat stemming from farm animals that have been treated with antibiotics? \newline In order to answer these questions health risk analysis is divided in risk assessment, risk communication and risk management. Thereby, health risk assessment estimates health risks to individuals or groups from hazardous exposures and decisions or actions which create them. In this regard specialized models and methods are used to quantify possible exposures and their associated health risks. Furthermore, health risk communication is used to edit the data of the risk assessment and present the resulting information to decision-makers and stakeholders. On the other hand health risk management consists of principles and methods for deciding between different alternatives or actions concerning exposures, health risks and/or their consequences. Each of these parts of health risk analysis is considered extensively in this book whereas many interesting questions (in theory as well as in applications) for operations researchers and management scientists are discussed. The book is divided in eight chapters. The first chapter provides an introduction and introduces basic risk models. Thereby, the core disciplines used in risk analysis are discussed, steps of a traditional risk analysis framework are presented and the causal structure of a risk (by answering the questions about its source, target, effect of concern and causal mechanism) is given. Moreover, the probability of binary events, hazard rate models, hazard functions, risk models with non-binary consequences and several models for risk management decision support are introduced. Chapter two is dedicated to basic techniques of risk assessment modelling. In this regard a concept for a conditional probability framework for risk calculations is given and some basic engineering models (compartmental flow models, Monte Carlo uncertainty models, stochastic transition models and discrete-event simulation models) are discussed. Furthermore, the key concepts of exposure assessment modelling are provided and a case study applying some of these techniques is presented. Chapter three deals with statistical risk modelling which provides approaches for describing and predicting probabilities of health responses on individual exposures and covariates. Thereby, statistical dose-response modelling is discussed extensively. In addition, the key ideas of advances in statistics and statistical computation, which led to an improvement in practice of statistical risk assessment, are summarized. At last, as in chapter two, a case study is presented, showing how modern statistical methods can be used in order to solve a statistical risk management decision problem. In chapter four techniques for modelling causal relations between a change in health risks and a given change in exposures are described. To this end criteria for causation are given and causal graph models are introduced. Moreover, it is shown how causal graphs can be used in risk analysis and attributable risks in causal graphs are discussed. Chapter five is the first of three chapters (together with chapter seven and eight) dealing with how Quantitative Risk Assessment (QRA) can be used to improve choices among risk management options. Furthermore topics like how to use QRA information to identify useful and effective risk management options are introduced through examples and via summaries of experimental evidence. Moreover, this chapter deals with Expected Utility (EU) theory, which justifies the use of optimization techniques for decision making when the objective function is known. Chapter six covers methods for choosing among risk profiles -- or, equivalently, cumulative probability distribution functions (CDFs) -- when the consequences of interest can be summarized and compared using a single numerical attribute. These methods provide ways to compare risk profiles and to eliminate those that should not come to pass. The results in this chapter can be used to construct explicit utility functions for single-attribute consequences so that the expected utilities of prospects can be calculated. In chapter seven the analysis to multi-attribute consequences and utility theory is extended and models of decision-making over time are introduced. Approaches for extending EU theory to multiple attributes, stakeholders, time periods and generations are extended. It also includes an approach to choosing among Markov or semi-Markov transition processes with multiple health states. Further topics, like attempts to extend expected theory to represent and prescribe preferences for risk equity, multi-person consequences, multi-criteria consequences, intergenerational welfare, and a variety of temporal prospects, are illustrated considering examples. Chapter eight considers risk management decision-making for multiple decision-makers and stakeholders. The causal, statistical, and simulation-based models and methods of analysis in chapter two to four and the decision-analytic models and methods of chapter five to seven can produce and structure risk and uncertainty information to inform and improve decisions. The risk communication and decision-process design approaches surveyed in this chapter bring such technical information to bear at points in decision processes where it can most clarify and improve analysis, deliberation, and risk management decisions. The book contains more than 500 pages and more than 650 references. It is well written and structured, and in particular, the extensive number of examples provided throughout the book give an excellent insight in the topic of health risk analysis.