A reflected feature space for CART

We present an algorithm for learning oblique decision trees, called HHCART(G). Our decision tree combines learning concepts from two classification trees, HHCART and Geometric Decision Tree (GDT). HHCART(G) is a simplified HHCART algorithm that uses linear structure in the training examples, captured by a modified GDT angle bisector, to define splitting directions. At each node, we reflect the training examples with respect to the modified angle bisector to align this linear structure with the coordinate axes. Searching axis parallel splits in this reflected feature space provides an efficient and effective way of finding oblique splits in the original feature space. Our method is much simpler than HHCART because it only considers one reflected feature space for node splitting. HHCART considers multiple reflected feature spaces for node splitting making it more computationally intensive to build. Experimental results show that HHCART(G) is an effective classifier, producing compact trees with similar or better results than several other decision trees, including GDT and HHCART trees.     

Documenting Pornography Use in America: A Comparative Analysis of Methodological Approaches

Estimates of pornography use in the United States range widely. We explore the reasons for the variation in such estimates among U.S. adults using data from four different recent nationally representative samples—each of which asked a different type of question about pornography use. We attribute the notable variation in estimates to differences in question wording and answer options, and assert that a survey question asking respondents about their most recent use of pornography minimizes recall bias and is better poised to assess the overall prevalence of pornography in a population than is the more common approach of asking respondents about their historical general-use pattern. When we privileged the most-recent-use approach, survey data from 2014 reveal that 46% of men and 16% of women between the ages of 18 and 39 intentionally viewed pornography in a given week. These numbers are notably higher than most previous population estimates employing different types of questions. The results have ramifications for methods of surveying sensitive self-reported behaviors and for contextualizing scholars’ claims as well as popular conversations about the reach and implications of pornography use in the United States.     

Using the Theory of Planned Behaviour to Understand Motivation to Register Births in Lombok,Indonesia

Despite government interest in promoting birth registration, more than a third of Indonesian children do not have a birth certificate, affecting the realisation of both their human and citizen rights. While barriers to registering children's births in Indonesia have been assessed, there is limited research on how communities perceive the importance of having a birth certificate. This study used focus group discussions to explore parental motivations around birth registration. The results of a thematic analysis found that perceived use of birth certificates, perceived control over the application process and social norms related to certificate ownership affect the intention to apply.     

The influence of FDA advisory information and black box warnings on individual use of prescription antidepressants

We examine how use of antidepressant medications is influenced by Food and Drug Administration (FDA) warnings about the increased risk of suicidality associated with pediatric antidepressant use. With individual-level data on antidepressant use from the Medical Expenditure Panel Survey, we consider whether consumer responses to FDA warnings differ among targeted (children) and non-targeted (adult) age-groups. Because the warning labels specifically mentioned new users, we examine separately the effects of the warnings on initiations of antidepressant therapy and on continued use of antidepressants. We find evidence consistent with reduced initiation of antidepressant use among the intended population of children, and that usage among children with more highly educated parents responded earlier to FDA information. However, we also find spillover effects of reduced initiation among the non-targeted population of adults. Overall, our results indicate that the FDA warning may have led consumers to perceive risks beyond those specifically mentioned.     

Linking an agency strategic review to increase knowledge management: San Francisco County Human Service Agency

Led by the agency director, the agency engaged in a Strategic Review, based on a comprehensive assessment of agency performance that identified strategies to improve organizational effectiveness through increased data-informed practice and knowledge management. The Strategic Review gathered information on staff perceptions, perceptions of external stakeholders, changing citywide and neighborhood demographics, policy mandates, and budget and workload issues. The need for the review was based upon multiple, substantial changes not addressed in the 2000 Strategic Plan, including the 2004 merger of the Department of Human Services and the Department of Aging and Adult Services, changes among the executive management team, transitions among key political entities, new policy mandates and changing budget allocations. This case study describes the Strategic Review process and content, summarizing key challenges and lessons related to addressing workload demands, fostering positive staff attitudes, balancing internal and external information needs, and integrating data use and planning processes across the agency.     

Statistical modeling for Bayesian extrapolation of adult clinical trial information in pediatric drug evaluation

Children represent a large underserved population of “therapeutic orphans,” as an estimated 80% of children are treated off‐label. However, pediatric drug development often faces substantial challenges, including economic, logistical, technical, and ethical barriers, among others. Among many efforts trying to remove these barriers, increased recent attention has been paid to extrapolation; that is, the leveraging of available data from adults or older age groups to draw conclusions for the pediatric population. The Bayesian statistical paradigm is natural in this setting, as it permits the combining (or “borrowing”) of information across disparate sources, such as the adult and pediatric data. In this paper, authored by the pediatric subteam of the Drug Information Association Bayesian Scientific Working Group and Adaptive Design Working Group, we develop, illustrate, and provide suggestions on Bayesian statistical methods that could be used to design improved pediatric development programs that use all available information in the most efficient manner. A variety of relevant Bayesian approaches are described, several of which are illustrated through 2 case studies: extrapolating adult efficacy data to expand the labeling for Remicade to include pediatric ulcerative colitis and extrapolating adult exposure‐response information for antiepileptic drugs to pediatrics.