A diagnostic for selecting the threshold in extreme value analysis

A new approach is suggested for choosing the threshold when fitting the Hill estimator of a tail exponent to extreme value data. Our method is based on an easily computed diagnostic, which in turn is founded directly on the Hill estimator itself, 'symmetrized' to remove the effect of the tail exponent but designed to emphasize biases in estimates of that exponent. The attractions of the method are its accuracy, its simplicity and the generality with which it applies. This generality implies that the technique has somewhat different goals from more conventional approaches, which are designed to accommodate the minor component of a postulated two-component Pareto mixture. Our approach does not rely on the second component being Pareto distributed. Nevertheless, in the conventional setting it performs competitively with recently proposed methods, and in more general cases it achieves optimal rates of convergence. A by-product of our development is a very simple and practicable exponential approximation to the distribution of the Hill estimator under departures from the Pareto distribution.     

Conditional approval: discussion points from the PSI conditional approval expert group

The European Agency for the Evaluation of Medicinal Products has recently completed the consultation of a draft guidance on how to implement conditional approval. This route of application is available for orphan drugs, emergency situations and serious debilitating or life-threatening diseases. Although there has been limited experience in implementing conditional approval to date, PSI (Statisticians in the Pharmaceutical Industry) sponsored a meeting of pharmaceutical statisticians with an interest in the area to discuss potential issues. This article outlines the issues raised and resulting discussions, based on the group's interpretation of the legislation. Conditional approval seems to fit well with the accepted regulatory strategy in HIV. In oncology, conditional approval may be most likely when (a) compelling phase II data are available using accepted clinical outcomes (e.g. progression/recurrence-free survival or overall survival) and Phase III has been planned or started, or (b) when data are available using a surrogate endpoint for clinical outcome (e.g. response rate or biochemical measures) from a single-arm study in rare tumours with high response, compared with historical data. The use of interim analyses in Phase III for supporting conditional approval raises some challenging issues regarding dissemination of information, maintenance of blinding, potential introduction of bias, ethics, switching, etc.     

The issue of consent in research that administers drugs of addiction to addicted persons

In addiction, impaired control over drug use raises questions about the capacity of addicted persons to consent to participate in research studies in which they are given their drug of addiction. We review the case for doing such research, and the arguments that addiction does, and does not, prevent addicted persons from consenting to such research. We argue for a more nuanced view that acknowledges that while in some situations addiction impairs decision-making capacity, it does not eliminate such capacity. We conclude with some suggestions for recruiting addicted subjects and designing experiments in ways to obtain free and informed consent.     

On the Addition and Comparison of Graphs Labeled with Stochastic Variables: Learnable Anatomical Catalogs

We provide an operator for the addition of a pair of graphs, labeled with continuous variables which are subject to stochastic variation. We also provide an operator for measuring dissimilarity between a pair of such graphs. We use such a representation and operators to model a collection of vascular anatomy which accounts for inter-individual variations in both branching structure and in vessel shape. The model may be incrementally acquired, and is thus a catalog of anatomy whose content may be learned. The model may be used in applications such as the reconstruction of vasculature in three-dimensions from x-ray images, which we briefly outline.     

Qualitative exploration of the nature of grief-related beliefs and expectations

Societal expectations of grief impact the experience of bereavement. The congruence of societal expectations with current scientific understanding of grief is unknown. Therefore two qualitative studies explored community perceptions of grief. In study one, three small focus groups (N = 9) examined grief-related expectations associated with hypothetical scenarios of bereavement. In study two, the impact of grief-related perceptions on the lived experience of bereavement for 11 individuals was explored through semi-structured interviews. Across both studies, elements of a traditional stage model view of grief were evident, with participants viewing emotional expression of grief as important. An avoidant coping style in the bereaved was considered problematic. Findings of study two suggested that grief-related beliefs may impact the bereavement experience via appraisal of the grief response and willingness to support bereaved individuals. The studies suggested that stage model assumptions in the beliefs of the general population persist, although there was a recognition of diversity in the grief response.     

Iowa Case Management for Rural Drug Abuse

OBJECTIVE: The purpose of this research was to evaluate the effectiveness of a comprehensive, strengths-based model of case management for clients in drug abuse treatment. METHOD: 503 volunteers from residential or intensive outpatient treatment were randomly assigned to one of three conditions of Iowa Case Management (ICM) plus treatment as usual (TAU), or to a fourth condition of TAU only. All were assessed at intake and followed at 3, 6, and 12 months. RESULTS: Clients in all four conditions significantly decreased substance use by 3 months after intake and maintained most gains over time. However, the addition of ICM to TAU did not improve substance use outcomes. CONCLUSION: Overall, the addition of case management did not significantly improve drug treatment as hypothesized by both researchers and clinicians. Some results were mixed, possibly due to the heterogeneous sample, wide range of case management activities, or difficulty retaining participants over time.     

Regional listening forums: an examination of the methodologies used by the child welfare league of America and lambda legal to highlight the experiences of LGBTQ youth in care

In 2002, the Child Welfare League of America (CWLA) and the Lambda Legal Defense and Education Fund began Fostering Transitions: CWLA/Lambda Joint Initiative to Support LGBTQ Youth and Adults Involved with the Child Welfare System. To document the needs of lesbian, gay, bisexual, transgender, and questioning (LGBTQ) youth, as well as identify strategies for systems improvement, initiative staff associated with the joint initiative conducted a series of Regional Listening Forums in 13 cities in the United States. More than 500 participants attended the forums, representing 22 states from every region in the country. Participants included former and current youth in care as well as the adults who work most closely with them. This article focuses on the methodologies on which the forums were developed and conducted. I realized that being gay is not my problem. It's their problem. I see it as a social disease. I try not to get involved in negative communities. But I do try to teach them. I'd rather teach them than ignore them. Otherwise, the ignorance will continue and nothing will ever be done about it.     

Aftermath of tragic events: the development and use of community support meetings on a university campus

Colleges and universities need to be prepared to address the psychological impact of tragedies on their campuses. In this article, the author describes the development and successful implementation of campus postvention services in the aftermath of college student deaths by suicide as well as by natural and accidental causes. The program has been well received and has gone a long way toward helping the campus community come together and heal after these types of tragedies. College officials adapted the program to address issues related to troubling national and international events as well local incidents, such as suicide attempts and students' loss of housing through fires. The author provides specific instructions so that readers can replicate the program on their own campuses, and additional materials are available upon request.     

Characteristics of social smoking among college students

Social smoking is a newly identified phenomenon in the young adult population that is poorly understood. We investigated differences in social smoking (smoking most commonly while partying or socializing) and other smoking within a convenience sample of college smokers (n = 351) from a large midwestern university. Results revealed that 70% of 351 current (past 30-day) smokers reported social smoking. No significant difference was found in motivation to quit between smoking groups. However, a significant difference was found between groups in confidence to quit, the number of days smoked, and the number of cigarettes smoked on those days. More social smokers than expected did not perceive themselves as smokers. Logistic regression analysis revealed that lower physical and psychological dependence and higher social support scores predicted social smoking.     

Individual-level modeling of the spread of influenza within households

A class of individual-level models (ILMs) outlined by R. Deardon et al., [Inference for individual level models of infectious diseases in large populations, Statist. Sin. 20 (2010), pp. 239–261] can be used to model the spread of infectious diseases in discrete time. The key feature of these ILMs is that they take into account covariate information on susceptible and infectious individuals as well as shared covariate information such as geography or contact measures. Here, such ILMs are fitted in a Bayesian framework using Markov chain Monte Carlo techniques to data sets from two studies on influenza transmission within households in Hong Kong during 2008 to 2009 and 2009 to 2010. The focus of this paper is to estimate the effect of vaccination on infection risk and choose a model that best fits the infection data.