Expertises croisées dans la dégénérescence maculaire liée à l'âge. Focus sur la physiopathologie, l'angiogenèse, les données pharmacologiques et cliniques [Review and expert opinion in age related macular degeneration. Focus on the pathophysiology, angiogenesis and pharmacological and clinical data].

Age related macular degeneration (AMD) is a pathological aging of the macula, brought about by the interaction of genetic and environmental factors. It induces geographic atrophy of the retina and/or choroidal neovascularization. In the latter, abnormal vessels develop from the choriocapillaris, with the involvement of VEGF (vascular endothelial growth factor). The VEGF family includes several factors, including VEGF-A, B, C, D, F and PlGF (placental growth factor). Their biological properties and their affinities to the VEGFR1, VEGFR2 and VEGFR3 receptors found on endothelial cells differ. Exudative AMD involves mainly VEGF-A and VEGF-R2. Anti-VEGF agents used in ophthalmology (ranibizumab, bevacizumab and aflibercept) are designed to primarily target this pathway. In vitro, all have sufficient affinity to their ligands. Their therapeutic efficacy must therefore be judged based on clinical criteria. In clinical practice, the minimum number of injections required for a satisfactory result appears to be comparable with all the three. The few available studies on therapeutic substitutions of anti-VEGF compounds suggest that some patients may benefit from substituting the anti-VEGF in cases of an unsatisfactory response to an initial molecule. Although local side effects, including increased risk of geographic atrophy, and systemic effects, including vascular accidents, have been suggested, these risks remain low, specially compared to the benefits of the treatment. Differences in safety between anti-VEGF are theoretically possible but unproven.

Textable: programmation visuelle pour l'analyse de données textuelles

TEXTABLE est un nouvel outil open source de programmation visuelle pour l'analyse de données textuelles. Les implications de la conception de ce logiciel du point de vue de l'interopérabilité et de la flexibilité sont abordées, ainsi que la question que son adéquation pour un usage pédagogique. Une brève introduction aux principes de la programmation visuelle pour l'analyse de données textuelles est également proposée.

Analyse des séjours hospitaliers : apport des données infirmières

Background: In Switzerland the aim of health insurance (LAMaI) is to control rising health costs. One method is to institute a prospective payment system (DRGs) for all Swiss hospitals in 2012, according to which hospital funding will be based only on medical and administrative data. Question: What is the contribution of nursing to the analysis of hospital stays? Method: On the basis of aIl patient hospital stays in the CHUV (Lausanne) during 2005 and 2006, we have compared a medico-administrative data model and a nursing data model. We used a logistic regression on the probability of patient exit. Results: The capacity of discrimination of the model is appreciably improved since the surface under curve O.C.R. passes from 0.712 with the casemix data and the rang of day to 0.785 if it's adds data on heaviness of care (pseudo R² respectively 0.096 and 0.152). Discussion: This approach provides evidence on the feasibility of using standards of care pathway allowing managers to analyse the organisation of patient care and securing a better estimate of hospital funding.

Breaking the panels: an application to the GDP per capita

This paper proposes a test statistic for the null hypothesis of panel stationarity that allows for the presence of multiple structural breaks. Two different speci¿cations are considered depending on the structural breaks affecting the individual effects and/or the time trend. The model is ¿exible enough to allow the number of breaks and their position to differ across individuals. The test is shown to have an exact limit distribution with a good ¿nite sample performance. Its application to a typical panel data set of real per capita GDP gives support to the trend stationarity of these series

Breaking the panels: an application to the GDP per capita

This paper proposes a test statistic for the null hypothesis of panel stationarity that allows for the presence of multiple structural breaks. Two different speci¿cations are considered depending on the structural breaks affecting the individual effects and/or the time trend. The model is ¿exible enough to allow the number of breaks and their position to differ across individuals. The test is shown to have an exact limit distribution with a good ¿nite sample performance. Its application to a typical panel data set of real per capita GDP gives support to the trend stationarity of these series

Jugements moraux et motivation à la lumière des données empiriques

Cet article contient un « tableau affectif », une analyse détaillée, largement basée sur des données empiriques, de la manière dont les gens jugent et se comportement moralement, où sont précisés les rôles respectifs des processus réflexifs et affectifs dans l'activité morale. Il est utile de distinguer différentes sortes de jugements, certains étant plus complexes que d'autres. Les « jugements sophistiqués » interviennent dans le cadre d'une activité rationnelle alors que les « jugements de valeur spontanés » sont une manière primitive et non réflexive d'évaluer le monde. Cette analyse permettra de prendre position dans le traditionnel débat sur l'internalisme et l'externalisme. Je défendrai l'idée que les normes et les jugements moraux que nous défendons n'exercent pas d'effet particulier sur nos actions si nous n'avons pas, par ailleurs, une attitude émotionnelle correspondante. Or cette dernière semble relever davantage de nos jugements spontanés.