Statistics in Medicine,2012年31(7):681-697 ISSN：1097-0258
[Feng, Ping] Sichuan Univ, W China Hosp, Inst Clin Trials, Chengdu, Sichuan, Peoples R China.;[Li, Xiao-Song] Sichuan Univ, W China Sch Publ Hlth, Chengdu, Sichuan, Peoples R China.;[Zhou, Xiao-Hua] Harbin Med Coll, Harbin, Peoples R China.;[Zhou, Xiao-Hua] Univ Washington, Sch Publ Hlth, Dept Biostat, Seattle, WA 98195 USA.;[Zhou, Xiao-Hua] Peking Univ, Beijing Int Ctr Math Res, Beijing 100871, Peoples R China.
[Li, Xiao-Song] Sichuan Univ, W China Sch Publ Hlth, Chengdu, Sichuan, Peoples R China.
causal inference<&wdkj&>generalized propensity score<&wdkj&>treatment effect<&wdkj&>Traditional Chinese Medicine (TCM)<&wdkj&>multiple treatment components
The propensity score method is widely used in clinical studies to estimate the effect of a treatment with two levels on patient's outcomes. However, due to the complexity of many diseases, an effective treatment often involves multiple components. For example, in the practice of Traditional Chinese Medicine (TCM), an effective treatment may include multiple components, e.g. Chinese herbs, acupuncture, and massage therapy. In clinical trials involving TCM, patients could be randomly assigned to either the treatment or control group, but they or their doctors may make different choices about which treatment component to use. As a result, treatment components are not randomly assigned. Rosenbaum and Rubin proposed the propensity score method for binary treatments, and Imbens extended their work to multiple treatments. These authors defined the generalized propensity score as the conditional probability of receiving a particular level of the treatment given the pre-treatment variables. In the present work, we adopted this approach and developed a statistical methodology based on the generalized propensity score in order to estimate treatment effects in the case of multiple treatments. Two methods were discussed and compared: propensity score regression adjustment and propensity score weighting. We used these methods to assess the relative effectiveness of individual treatments in the multiple-treatment IMPACT clinical trial. The results reveal that both methods perform well when the sample size is moderate or large. Copyright (c) 2011 John Wiley & Sons, Ltd.
digital main control room;cognitive reliability;human factors issues;human errors;bayesian network
Currently, there is a trend in nuclear power plants (NPPs) toward introducing digital and computer technologies into main control rooms (MCRs). Safe generation of electric power in NPPs requires reliable performance of cognitive tasks such as fault detection, diagnosis, and response planning. The digitalization of MCRs has dramatically changed the whole operating environment, and the ways operators interact with the plant systems. If the design and implementation of the digital technology is incompatible with operators' cognitive characteristics, it may have negative effects on operators' cognitive reliability. Firstly, on the basis of three essential prerequisites for successful cognitive tasks, a causal model is constructed to reveal the typical human performance issues arising from digitalization. The cognitive mechanisms which they impact cognitive reliability are analyzed in detail. Then, Bayesian inference is used to quantify and prioritize the influences of these factors. It suggests that interface management and unbalanced workload distribution have more significant impacts on operators' cognitive reliability.