Drug Utilization Studies and Data Registries in Primary Care.

This article reviews the methodology of the studies on drug utilization with particular emphasis on primary care. Population based studies of drug inappropriateness can be done with microdata from Health Electronic Records and e-prescriptions. Multilevel models estimate the influence of factors affecting the appropriateness of drug prescription at different hierarchical levels: patient, doctor, health care organization and regulatory environment. Work by the GIUMAP suggest that patient characteristics are the most important factor in the appropriateness of prescriptions with significant effects at the general practicioner level.

Técnicas de Data e Text Mining Para Anotação de Um Arquivo Digital

O presente trabalho cujo Título é técnicas de Data e Text Mining para a anotação dum Arquivo Digital, tem como objectivo testar a viabilidade da utilização de técnicas de processamento automático de texto para a anotação das sessões dos debates parlamentares da Assembleia da República de Portugal. Ao longo do trabalho abordaram-se conceitos como tecnologias de descoberta do conhecimento (KDD), o processo da descoberta do conhecimento em texto, a caracterização das várias etapas do processamento de texto e a descrição de algumas ferramentas open souce para a mineração de texto. A metodologia utilizada baseou-se na experimentação de várias técnicas de processamento textual utilizando a open source R/tm. Apresentam-se, como resultados, a influência do pré-processamento, tamanho dos documentos e tamanhos dos corpora no resultado do processamento utilizando o algoritmo knnflex.

The Impact of Industry Collaboration on Academic Research Output: A Dynamic Panel Data Analysis

The aim of this paper is to analyse the impact of university knowledge and technology transfer activities on academic research output. Specifically, we study whether researchers with collaborative links with the private sector publish less than their peers without such links, once controlling for other sources of heterogeneity. We report findings from a longitudinal dataset on researchers from two engineering departments in the UK between 1985 until 2006. Our results indicate that researchers with industrial links publish significantly more than their peers. Academic productivity, though, is higher for low levels of industry involvement as compared to high levels.

Pliensbachian magnetostratigraphy: new data from Paris Basin (France)

Sampling of an industrial drill string from the northeastern Paris Basin (Montcornet, France) provides early Jurassic magnetostratigraphic data coupled with biochronological control. About 375 paleomagnetic samples were obtained from a 145 m thick series of Pliensbachian rocks. A composite demagnetization thermal up to 300 C and an alternating field up to 80 mT were used to separate the magnetic components. A low unblocking temperature component (<250degreesC) with an inclination of about 64 is interpreted as a present-day field overprint. The characteristic remanent component with both normal and reversed antipodal directions was isolated between 5 and 50 mT. Twenty-nine polarity intervals were recognized. Correlation of these new results from the Paris Basin with data from the Breggia Gorge section (Ticino, southern Alps, Switzerland), which is generally considered as the reference section for Pliensbachian magnetostratigraphy, reveals almost identical patterns of magnetic polarity reversals. However, the correlation implies significant paleontological age discrepancies. Revised age assignments of biostratigraphic data of Breggia as well as an objective evaluation of the uncertainties on zonal boundaries in both Breggia and Moncornet resolve the initial discrepancies between magnetostratigraphic correlations and biostratigraphic ages. Hence, the sequence of magnetic reversals is significantly strengthened and the age calibration is notably improved for the Pliensbachian, a stage for which sections combining adequate magnetic signal and biostratigraphic constraints are still very few. (C) 2002 Elsevier Science B.V. All rights reserved.