Pruning regression trees for censored survival data: the recpam approach

RECPAM is a methodology, implemented in a computer program of the same name, for the construction of tree-structured models in Biostatistics. In this work we present algorithms for pruning and amalgamating terminal nodes of a tree, within the RECPAM approach. These algorithms construct sequences of nested models and calculate at each step the AIC of the corresponding model and correct significance levels, according to Gabriel's theory of Simultaneous Test Procedures. As an example, the analysis of data from clinical trials involving patients with Small Cell Carcinoma of the Lung is presented.     

Safe and shealthy integration into semiskilled jobs: does gender matter?

Women report fewer injuries than men when they are young, but report a higher proportion of professional diseases later in life (35 years +). Understanding early work conditions that lead to postponed health outcomes is necessary if we are to prevent women's injury as well as men's. Introduction to work and safe integration programs are often put forward to prevent injuries among youth. This paper aims to illustrate some differences in the job introduction / integration process for men and women and to discuss some gender-based Occupational Health and Safety (OHS) implications. Data come from two waves of interviews with 31 students enrolled in vocational training for a semiskilled trade. Results demonstrated differences between men and women: trades chosen, OHS complaints, supervising patterns, integration to work, etc. Women often choose customer service jobs, have less formal supervision and are often left to themselves when learning to do their job. Men declared more accidents while women reported more musculoskeletal symptoms. Findings from this research suggest young women's OHS issues should not be trivialized and that specific youth prevention programs for sectors where women are more present, such as customer service, should be developed.     

Associations Between Maternal Postpartum Depression,Infant Social Behavior With a Stranger,and Infant Cognitive Development

Maternal postpartum depression (PPD) has been found to be related to infant social withdrawal during mother–infant interaction, and this may spill over on infant interactive behavior in other social contexts and impact infant psychosocial development. We investigated whether PPD was associated with infant social withdrawal during interaction with a tester in a psychological test situation and whether infant social withdrawal in the test situation mediated the association between PPD and infant cognitive scores reported in a previous study. Participants were 28 PPD dyads and 41 control dyads. We assessed infant social behavior and cognitive development with the Alarm Distress Baby Scale and the cognitive scale of the Bayley Scales of Infant and Toddler Development, third edition, at four months. More symptoms of maternal depression were associated with more infant social withdrawal. The association between maternal depressive symptoms and cognitive scores was at most partially mediated by infant social withdrawal in the test situation (<29.6%). Our results add to the existing literature on the effects of PPD on infant social behavior in other contexts than the one constituted by the mother. More research is needed to shed light on the mechanisms through which PPD impacts infant cognitive development.     

A Communication Perspective on Video Lottery Terminals

Video Lottery Terminals (VLT) are associated with pathological gambling and with most of the requests for help in combating gambling addiction. Embeddedness of a person in his or her social network is among the communicational factors that may help explain this phenomenon. To verify this, we compared ego networks of VLT gamblers, of gamblers of games with low request for help and of VLT gamblers in treatment (n = 90). The networks of regular VLT gamblers are small and dense and offer little social support. Gamblers in treatment also have small networks, but they are less dense, have more components and offer more social support. Networks of gamblers with low requests for assistance are approximately twice the size as those of VLT gamblers, are sparser and offer more companionship. In conclusion, the VLT gambler is not an isolated individual, but rather an individual ‘shut-in’ a small network of tightly knitted relationships.