Type I error rates from likelihood‐based repeated measures analyses of incomplete longitudinal data

The last observation carried forward (LOCF) approach is commonly utilized to handle missing values in the primary analysis of clinical trials. However, recent evidence suggests that likelihood‐based analyses developed under the missing at random (MAR) framework are sensible alternatives. The objective of this study was to assess the Type I error rates from a likelihood‐based MAR approach – mixed‐model repeated measures (MMRM) – compared with LOCF when estimating treatment contrasts for mean change from baseline to endpoint (Δ). Data emulating neuropsychiatric clinical trials were simulated in a 4 × 4 factorial arrangement of scenarios, using four patterns of mean changes over time and four strategies for deleting data to generate subject dropout via an MAR mechanism. In data with no dropout, estimates of Δ and SE_Δ from MMRM and LOCF were identical. In data with dropout, the Type I error rates (averaged across all scenarios) for MMRM and LOCF were 5.49% and 16.76%, respectively. In 11 of the 16 scenarios, the Type I error rate from MMRM was at least 1.00% closer to the expected rate of 5.00% than the corresponding rate from LOCF. In no scenario did LOCF yield a Type I error rate that was at least 1.00% closer to the expected rate than the corresponding rate from MMRM. The average estimate of SE_Δ from MMRM was greater in data with dropout than in complete data, whereas the average estimate of SE_Δ from LOCF was smaller in data with dropout than in complete data, suggesting that standard errors from MMRM better reflected the uncertainty in the data. The results from this investigation support those from previous studies, which found that MMRM provided reasonable control of Type I error even in the presence of MNAR missingness. No universally best approach to analysis of longitudinal data exists. However, likelihood‐based MAR approaches have been shown to perform well in a variety of situations and are a sensible alternative to the LOCF approach. MNAR methods can be used within a sensitivity analysis framework to test the potential presence and impact of MNAR data, thereby assessing robustness of results from an MAR method. Copyright © 2004 John Wiley & Sons, Ltd.     

保障奥运食品安全 完善食品安全可追溯管理

近年来，食品安全成为全球热点问题之一，为解决食品安全问题，食品可追溯技术应运而生，只要一个条形码，就可追溯到食品的出产地、生长地甚至基因情况，食品有了自己的“身份证”。为保障奥运期间的食品安全，北京应加强食品安全管理，供应奥运食品将全部实行可追溯制度，用科技手段来为奥运会护航。     

Parent‐Adolescent Communication About Sex: Retrospective Reports by Latino College Students

Latina female (n= 97) and Latino male (n= 69) college students (M age = 21.4 years) completed self‐report surveys regarding family of origin experiences, including sexual communication with parents while growing up. Latino parents of this comparatively highly educated sample tended to use direct rather than indirect strategies for communicating about sexuality with their children. Young women reported higher levels of sexual communication with mothers while growing up than did young men, and respondents reported less communication with fathers than mothers. Among young women, sexual communication with mother was positively associated with non‐Mexican origin and negatively associated with having older brothers living at home. In contrast, maternal education was positively associated with mother‐son communication about sex. Paternal education and the absence of older brothers positively predicted communication with both sons and daughters. The analyses provide novel information regarding sexual communication in Latino families and suggest directions for future research.     

Choice of the primary analysis in longitudinal clinical trials

Missing data, and the bias they can cause, are an almost ever‐present concern in clinical trials. The last observation carried forward (LOCF) approach has been frequently utilized to handle missing data in clinical trials, and is often specified in conjunction with analysis of variance (LOCF ANOVA) for the primary analysis. Considerable advances in statistical methodology, and in our ability to implement these methods, have been made in recent years. Likelihood‐based, mixed‐effects model approaches implemented under the missing at random (MAR) framework are now easy to implement, and are commonly used to analyse clinical trial data. Furthermore, such approaches are more robust to the biases from missing data, and provide better control of Type I and Type II errors than LOCF ANOVA. Empirical research and analytic proof have demonstrated that the behaviour of LOCF is uncertain, and in many situations it has not been conservative. Using LOCF as a composite measure of safety, tolerability and efficacy can lead to erroneous conclusions regarding the effectiveness of a drug. This approach also violates the fundamental basis of statistics as it involves testing an outcome that is not a physical parameter of the population, but rather a quantity that can be influenced by investigator behaviour, trial design, etc. Practice should shift away from using LOCF ANOVA as the primary analysis and focus on likelihood‐based, mixed‐effects model approaches developed under the MAR framework, with missing not at random methods used to assess robustness of the primary analysis. Copyright © 2004 John Wiley & Sons, Ltd.     

实数域及四元数除环上有平方根的方阵

给出了实数域及实四元数除环上方阵有平方根的充分必要条件．     

Marginal and association regression models for longitudinal binary data with drop‐outs: A likelihood‐based approach

Longitudinal data often contain missing observations, and it is in general difficult to justify particular missing data mechanisms, whether random or not, that may be hard to distinguish. The authors describe a likelihood‐based approach to estimating both the mean response and association parameters for longitudinal binary data with drop‐outs. They specify marginal and dependence structures as regression models which link the responses to the covariates. They illustrate their approach using a data set from the Waterloo Smoking Prevention Project They also report the results of simulation studies carried out to assess the performance of their technique under various circumstances.     

论科技期刊的品牌资本

科技期刊的品牌资本是其生存和发展的关键因素,品牌资本体现了社会效益与经济效益的同一性。品牌资本的价值回归是一个缓慢但却是相当稳定的过程。通过抽样调查和方差分析,定量说明了上述论点的正确性。     

Potential exposures and risks from beryllium-containing products.

Beryllium is the strongest of the lightweight metals. Used primarily in military applications prior to the end of the Cold War, beryllium is finding new applications in many commercial products, including computers, telecommunication equipment, and consumer and automotive electronics. The use of beryllium in nondefense consumer applications is of concern because beryllium is toxic. Inhalation of beryllium dust or vapor causes a chronic lung disease in some individuals at concentrations as low as 0.01 microg/m3 in air. As beryllium enters wider commerce, it is prudent to ask what risks this might present to the general public and to workers downstream of the beryllium materials industry. We address this question by evaluating the potential for beryllium exposure from the manufacturing, use, recycle, and disposal of beryllium-containing products. Combining a market study with a qualitative exposure analysis, we determine which beryllium applications and life cycle phases have the largest exposure potential. Our analysis suggests that use and maintenance of the most common types of beryllium-containing products do not result in any obvious exposures of concern, and that maintenance activities result in greater exposures than product use. Product disposal has potential to present significant individual risks, but uncertainties concerning current and future routes of product disposal make it difficult to be definitive. Overall, additional exposure and dose-response data are needed to evaluate both the health significance of many exposure scenarios, and the adequacy of existing regulations to protect workers and the public. Although public exposures to beryllium and public awareness and concern regarding beryllium risks are currently low, beryllium risks have psychometric qualities that may lead to rapidly heightened public concern.     

Costs and Rewards of Children: The Effects of Becoming a Parent on Adults' Lives

How do new parents differ from their childless counterparts in social and psychological resources, daily strains, and psychological well‐being? Using a nationally representative panel of 1,933 adults who were childless at the first interview, we compare 6 indicators of adults' lives for those who became parents and those remaining childless several years later, controlling for earlier states. Becoming a parent is both detrimental and rewarding. With the exception of social integration, which is greater for all groups of new parents compared with their childless counterparts, the effects of parental status on adults' lives vary markedly by gender and marital status. Unmarried parents report lower self‐efficacy and higher depression than their childless counterparts. Married mothers' lives are marked by more housework and more marital conflict but less depression than their childless counterparts. Parental status has little influence on the lives of married men.     

Valuation of childhood risk reduction: the importance of age, risk preferences, and perspective.

This article explores two problems analysts face in determining how to estimate values for children's health and safety risk reductions. The first addresses the question: Do willingness‐to‐pay estimates for health risk changes differ across children and adults and, if so, how? To answer this question, the article first examines the potential effects of age and risk preferences on willingness to pay. A summary of the literature reporting empirical evidence of differences between willingness to pay for adult health and safety risk reductions and willingness to pay for health and safety risk reductions in children is also provided. The second dimension of the problem is a more fundamental issue: Whose perspective is relevant when valuing children's health effects—society's, children's, adults‐as‐children, or parents'? Each perspective is considered, followed ultimately by the conclusion that adopting a parental perspective through an intrahousehold allocation model seems closest to meeting the needs of the estimation problem at hand. A policy example in which the choice of perspective affects the outcome of a regulatory benefit‐cost analysis rounds out the article and emphasizes the importance of perspective.