Using recurrent time-to-event models with multinomial outcomes to generate toxicity profiles

Most clinical studies, which investigate the impact of therapy simultaneously, record the frequency of adverse events in order to monitor safety of the intervention. Study reports typically summarise adverse event data by tabulating the frequencies of the worst grade experienced but provide no details of the temporal profiles of specific types of adverse events. Such 'toxicity profiles' are potentially important tools in disease management and in the assessment of newer therapies including targeted treatments and immunotherapy where different types of toxicity may be more common at various times during long-term drug exposure. Toxicity profiles of commonly experienced adverse events occurring due to exposure to long-term treatment could assist in evaluating the costs of the health care benefits of therapy. We show how to generate toxicity profiles using an adaptation of the ordinal time-to-event model comprising of a two-step process, involving estimation of the multinomial response probabilities using multinomial logistic regression and combining these with recurrent time to event hazard estimates to produce cumulative event probabilities for each of the multinomial adverse event response categories. Such a model permits the simultaneous assessment of the risk of events over time and provides cumulative risk probabilities for each type of adverse event response. The method can be applied more generally by using different models to estimate outcome/response probabilities. The method is illustrated by developing toxicity profiles for three distinct types of adverse events associated with two treatment regimens for patients with advanced breast cancer.     

Prior distributions for variance parameters in a sparse-event meta-analysis of a few small trials

In rare diseases, typically only a small number of patients are available for a randomized clinical trial. Nevertheless, it is not uncommon that more than one study is performed to evaluate a (new) treatment. Scarcity of available evidence makes it particularly valuable to pool the data in a meta-analysis. When the primary outcome is binary, the small sample sizes increase the chance of observing zero events. The frequentist random-effects model is known to induce bias and to result in improper interval estimation of the overall treatment effect in a meta-analysis with zero events. Bayesian hierarchical modeling could be a promising alternative. Bayesian models are known for being sensitive to the choice of prior distributions for between-study variance (heterogeneity) in sparse settings. In a rare disease setting, only limited data will be available to base the prior on, therefore, robustness of estimation is desirable. We performed an extensive and diverse simulation study, aiming to provide practitioners with advice on the choice of a sufficiently robust prior distribution shape for the heterogeneity parameter. Our results show that priors that place some concentrated mass on small τ values but do not restrict the density for example, the Uniform(−10, 10) heterogeneity prior on the log(τ²) scale, show robust 95% coverage combined with less overestimation of the overall treatment effect, across varying degrees of heterogeneity. We illustrate the results with meta-analyzes of a few small trials.     

Critical Assessment of the Foundations of Power Transmission and Distribution Reliability Metrics and Standards

The U.S. federal government regulates the reliability of bulk power systems, while the reliability of power distribution systems is regulated at a state level. In this article, we review the history of regulating electric service reliability and study the existing reliability metrics, indices, and standards for power transmission and distribution networks. We assess the foundations of the reliability standards and metrics, discuss how they are applied to outages caused by large exogenous disturbances such as natural disasters, and investigate whether the standards adequately internalize the impacts of these events. Our reflections shed light on how existing standards conceptualize reliability, question the basis for treating large‐scale hazard‐induced outages differently from normal daily outages, and discuss whether this conceptualization maps well onto customer expectations. We show that the risk indices for transmission systems used in regulating power system reliability do not adequately capture the risks that transmission systems are prone to, particularly when it comes to low‐probability high‐impact events. We also point out several shortcomings associated with the way in which regulators require utilities to calculate and report distribution system reliability indices. We offer several recommendations for improving the conceptualization of reliability metrics and standards. We conclude that while the approaches taken in reliability standards have made considerable advances in enhancing the reliability of power systems and may be logical from a utility perspective during normal operation, existing standards do not provide a sufficient incentive structure for the utilities to adequately ensure high levels of reliability for end‐users, particularly during large‐scale events.     

Modeling Repeatable Events Using Discrete‐Time Data: Predicting Marital Dissolution

I join two methodologies by illustrating the application of multilevel modeling principles to hazard‐rate models with an emphasis on procedures for discrete‐time data that contain repeatable events. I demonstrate this application using data taken from the 1995 National Survey of Family Growth (NSFG) to ascertain the relationship between multiple covariates and risk of subsequent marital dissolution. I consider both fixed‐ and random‐effects versions of the multilevel model, as well as a Generalized Estimating Equation alternative to estimating random effects. I compare results obtained from the various estimators, noting why differences occur, and recommend when to choose the various alternatives. I also provide a set of SAS and STATA programs that can be used to analyze the NSFG data.     

Power Law Selection Model for Repairable Systems

A repairable system, under minimal repair, is usually modeled according to a Non-Homogeneous Poisson Process (NHPP) assuming a Power Law intensity function. A traditional approach considers iid NHPPs in order to conduct a statistical analysis based on a sample of systems. However, systems might be heterogeneous due to unmeasured variables such as age, suppliers, and so on. In order to verify this assumption a frequentist approach is proposed in this article. Some possible model scenarios considering different systems heterogeneity are compared using likelihood ratio tests and information criteria. Real data sets illustrate the proposed methodology.     

Family-Like Group in a Correctional Institution

This paper examines a particular residential group within a correctional treatment institution. The examination focuses on the family-likeness of this group. The concept of family-like as it exists in the literature is described and the features of family-likeness in the group under discussion are outlined. A comparison is made to other residential groups as well as to other groups described as family-like.     

Cross cultural differences in decisions from experience: Evidence from Denmark,Israel, and Taiwan

Three studies are presented that compare decisions from experience in Denmark, Israel, and Taiwan. They focus on two change-related cultural differences suggested by previous research on dialectical vs. analytic approach to thinking. The first implies that East Asians are more likely to change their behavior over time (i.e., are less consistent), the second that they expect more changes in the environment. The results show that the “less consistency in the East” hypothesis has a high predictive value. This hypothesis accurately predicts a behavioral pattern that was documented in all three studies, as well as a non-trivial effect of limited feedback in Study 3: When feedback was limited to the obtained payoff, the participants from Taiwan exhibited less risk aversion than the Israeli. Analysis of the “expecting more changes in the East” hypothesis reveals mixed results. This hypothesis was supported in Study 2, which examined relatively complex multi-alternative multi-outcome tasks, but not in Studies 1 and 3, which examined simple two-alternative two-outcome choice tasks. A possible explanation for the different predictive value of the two examined hypotheses is discussed.     

梁实秋事件：抗战议题设置与“学术民主”的缺失

梁实秋事件得出的教训,除了议题设置一方没严格按“学术民主”的程序出牌外,其在大众文化传播学意义上的教训必须深刻检讨。这就是：在大众媒介的强势宣传之下,如果掌控舆论强势的一方对不同意见的另一方面施加强大的政治情绪威慑力,并形成指导性主流压力,那么弱势的一方就会感到恐惧和孤独;由于支持他的人越来越少,于是他就只能面临两种选择：一是被迫放弃自己的意见,向主流意见屈服,一是继续孤独、顽抗下去,最终落荒而逃。     

论张说碑志的新变

在张说现存碑志中,官修碑志为奉敕为三品以上的重臣或皇亲国戚而作,《姚崇碑》为其最为特别之作。该碑在类型化的颂美之外,采用异常简写碑主族出、世系与乡邑,不写亡妻,运用皮里阳秋的笔法暗讽姚崇的孝心等方式表达“微言大义”,使其碑志由注重实用性向偏重文学性转变。私修碑志则是应朋友、亲戚、同僚之请而作,《贞节君碣》为其早年的代表作。该碑运用小说笔法、选择典型细节等方式,使模式化的碑志逐步成为生动形象的人物传记。这些变化体现了张说“实行”“素心”的创作理念,展现出文学性、个性化的特点。     

A temporal correlation function for the association between patient‐reported outcome and clinical endpoints

For clinical trials with multiple endpoints, the primary interest is usually to evaluate the relationship of these endpoints and treatment interventions. Studying the correlation of two clinical trial endpoints can also be of interests. For example, the association between patient‐reported outcome and clinically assessed endpoint could answer important research questions and also generate interesting hypothesis for future research. However, it is not straightforward to quantify such association. In this article, we proposed a multiple event approach to profile such association with a temporal correlation function, visualized by a correlation function plot over time with a confidence band. We developed this approach by extending the existing methodology in recurrent event literature. This approach was shown to be generally unbiased and could be a useful tool for data visualization and inference. We demonstrated the use of this method with data from a real clinical trial. Although this approach was developed to evaluate the association between patient‐reported outcome and adverse events, it can also be used to evaluate the association of any two endpoints that can be translated to time‐to‐event endpoints. Copyright © 2015 John Wiley & Sons, Ltd.