Locally Efficient Estimation in Censored Data Models: Theory and Examples

In many applications the observed data can be viewed as a censored high dimensional full data random variable X. By the curve of dimensionality it is typically not possible to construct estimators that are asymptotically efficient at every probability distribution in a semiparametric censored data model of such a high dimensional censored data structure. We provide a general method for construction of one-step estimators that are efficient at a chosen submodel of the full-data model, are still well behaved off this submodel and can be chosen to always improve on a given initial estimator. These one-step estimators rely on good estimators of the censoring mechanism and thus will require a parametric or semiparametric model for the censoring mechanism. We present a general theorem that provides a template for proving the desired asymptotic results. We illustrate the general one-step estimation methods by constructing locally efficient one-step estimators of marginal distributions and regression parameters with right-censored data, current status data and bivariate right-censored data, in all models allowing the presence of time-dependent covariates. The conditions of the asymptotics theorem are rigorously verified in one of the examples and the key condition of the general theorem is verified for all examples.

Clinical research projects at a German medical faculty: follow-up from ethical approval to publication and citation by others

BACKGROUND: Only data of published study results are available to the scientific community for further use such as informing future research and synthesis of available evidence. If study results are reported selectively, reporting bias and distortion of summarised estimates of effect or harm of treatments can occur. The publication and citation of results of clinical research conducted in Germany was studied. METHODS: The protocols of clinical research projects submitted to the research ethics committee of the University of Freiburg (Germany) in 2000 were analysed. Published full articles in several databases were searched and investigators contacted. Data on study and publication characteristics were extracted from protocols and corresponding publications. RESULTS: 299 study protocols were included. The most frequent study design was randomised controlled trial (141; 47%), followed by uncontrolled studies (61; 20%), laboratory studies (30; 10%) and non-randomised studies (29; 10%). 182 (61%) were multicentre studies including 97 (53%) international collaborations. 152 of 299 (51%) had commercial (co-)funding and 46 (15%) non-commercial funding. 109 of the 225 completed protocols corresponded to at least one full publication (total 210 articles); the publication rate was 48%. 168 of 210 identified publications (80%) were cited in articles indexed in the ISI Web of Science. The median was 11 citations per publication (range 0-1151). CONCLUSIONS: Results of German clinical research projects conducted are largely underreported. Barriers to successful publication need to be identified and appropriate measures taken. Close monitoring of projects until publication and adequate support provided to investigators may help remedy the prevailing underreporting of research.

Haematologic data, iron parameters and molecular findings in two new cases of iron-refractory iron deficiency anaemia

Matriptase-2 (Tmprss6), a type II transmembrane serine protease, has an essential role in iron homoeostasis as a hepcidin regulator. Recently, patients with TMPRSS6 mutations and suffering from iron-refractory iron deficiency anaemia (IRIDA) have been reported. We describe two new cases of IRIDA, one patient of Swiss origin and the second of Italian origin. The first case results from a large deletion of 1054 nucleotides corresponding to an in frame deletion of 30 amino acid residues in the low-density lipoprotein receptor-1/-2 (LDLR-1/-2) domains and from a missense mutation in CUB1 (S304L). In the second case, a homozygous G-->C mutation in the last nucleotide of exon 15 and which modified the consensus sequence of the 5' splice donor site of intron 15 (AGgt-->ACgt) was identified. Both patients had a high hepcidin level and low serum iron and transferrin saturation compared to age-matched controls. Continuous perfusion of i.v. iron 4 h/d x 5 d in the first case resulted in a significant rise in haemoglobin. These new cases of IRIDA illustrate the importance of LDLR-1/-2 and CUB1 domains in matriptase-2 function as well as the role of matriptase-2 in hepcidin regulation. Furthermore a deletional form of TMPRSS6 (in LDLR-1/-2 domains) resulting in IRIDA is described for the first time. These cases reinforce the belief that patients suffering from IRIDA have no specific geographical or ethnic distribution and are sporadic secondary to different mutations of the matriptase-2 gene.

EU Data Protection Law and Targeted Advertising: Consent and the Cookie Monster - Tracking the crumbs of online user behaviour

This article provides a holistic legal analysis of the use of cookies in Online Behavioural Advertising. The current EU legislative framework is outlined in detail, and the legal obligations are examined. Consent and the debates surrounding its implementation form a large portion of the analysis. The article outlines the current difficulties associated with the reliance on this requirement as a condition for the placing and accessing of cookies. Alternatives to this approach are explored, and the implementation of solutions based on the application of the Privacy by Design and Privacy by Default concepts are presented. This discussion involves an analysis of the use of code and, therefore, product architecture to ensure adequate protections.

Casting materials and their application in research and teaching

From a biological point of view, casting refers to filling of anatomical and/or pathological spaces with extraneous material that reproduces a three-dimensional replica of the space. Casting may be accompanied by additional procedures such as corrosion, in which the soft tissue is digested out, leaving a clean cast, or the material may be mixed with radiopaque substances to allow x-ray photography or micro computed topography (µCT) scanning. Alternatively, clearing of the surrounding soft tissue increases transparency and allows visualization of the casted cavities. Combination of casting with tissue fixation allows anatomical dissection and didactic surgical procedures on the tissue. Casting materials fall into three categories namely, aqueous substances (India ink, Prussian blue ink), pliable materials (gelatins, latex, and silicone rubber), or hard materials (methyl methacrylates, polyurethanes, polyesters, and epoxy resins). Casting has proved invaluable in both teaching and research and many phenomenal biological processes have been discovered through casting. The choice of a particular material depends inter alia on the targeted use and the intended subsequent investigative procedures, such as dissection, microscopy, or µCT. The casting material needs to be pliable where anatomical and surgical manipulations are intended, and capillary-passable for ultrastructural investigations.