More on Testing for Validity Instead of Looking for It

Using Monte Carlo simulations and reanalyzing the data of a validation study of the AEIM emotional intelligence test, we demonstrated that an atheoretical approach and the use of weak statistical procedures can result in biased validity estimates. These procedures included stepwise regression-and the general case of failing to include important theoretical controls-extreme scores analysis, and ignoring heteroscedasticity as well as measurement error. The authors of the AEIM test responded by offering more complete information about their analyses, allowing us to further examine the perils of ignoring theory and correct statistical procedures. In this paper we show with extended analyses that the AEIM test is invalid.

Diagnostic value of MRI for the detection of suspected placental invasion: Does it depend on observers' experience?

Purpose: To evaluate the diagnostic value of specific MR features for detection of suspected placental invasion according to observers' experience.Methods and Materials: Our study population included 25 pregnant women (mean age 35.16) investigated by prenatal MRI. In twelve out of them placental invasion was histopathologically proven, the 13 other women (52%) without placental invasion served as control group. Multiplanar T1- and T2-weighted sequences had been performed mostly without IV contrast injection (1.5 T). MR examinations of the two groups were rendered anonymous, mixed, then independently and retrospectively reviewed by two senior and two junior radiologists in view of 8 MR features indicating placentar invasion including the degree. Results were compared with surgical diagnosis (placenta normal/increta/accreta/percreta). Interobserver agrement between senior and junior readers were calculated. Stepwise logistic regression and receiver operating (ROC) curvers were performed.Results: Demographics between the two groups were not statistically different. Overall sensitivity and specificity for detecting placentar invasion was 90.9% and 75.0% for senior readers, and 81.8% and 61.8% for junior readers respectively. The most significant MR features indicating placentar invasion were T2 hypointense placental bands, followed by placenta praevia, focally interrupted myometrial border, posterior placental insertion, and heterogeneous placental signal. For each of the evaluated MR features the interobserver agreement kappa between the two senior readers was superior than that between the junior readers, ranging from bad (<0.4) to good (0.4-0.75).Conclusions: MRI can be a reliable and reproducible tool for detection of suspected placentar invasion, however very variable according to the observers' experience.

Development of an erythropoietin prescription simulator to improve abilities for the prescription of erythropoietin stimulating agents: is it feasible?

BACKGROUND: The increasing use of erythropoietins with long half-lives and the tendency to lengthen the administration interval to monthly injections call for raising awareness on the pharmacokinetics and risks of new erythropoietin stimulating agents (ESA). Their pharmacodynamic complexity and individual variability limit the possibility of attaining comprehensive clinical experience. In order to help physicians acquiring prescription abilities, we have built a prescription computer model to be used both as a simulator and education tool. METHODS: The pharmacokinetic computer model was developed using Visual Basic on Excel and tested with 3 different ESA half-lives (24, 48 and 138 hours) and 2 administration intervals (weekly vs. monthly). Two groups of 25 nephrologists were exposed to the six randomised combinations of half-life and administration interval. They were asked to achieve and maintain, as precisely as possible, the haemoglobin target of 11-12 g/dL in a simulated naïve patient. Each simulation was repeated twice, with or without randomly generated bleeding episodes. RESULTS: The simulation using an ESA with a half-life of 138 hours, administered monthly, compared to the other combinations of half-lives and administration intervals, showed an overshooting tendency (percentages of Hb values > 13 g/dL 15.8 ± 18.3 vs. 6.9 ± 12.2; P < 0.01), which was quickly corrected with experience. The prescription ability appeared to be optimal with a 24 hour half-life and weekly administration (ability score indexing values in the target 1.52 ± 0.70 vs. 1.24 ± 0.37; P < 0.05). The monthly prescription interval, as suggested in the literature, was accompanied by less therapeutic adjustments (4.9 ± 2.2 vs. 8.2 ± 4.9; P < 0.001); a direct correlation between haemoglobin variability and number of therapy modifications was found (P < 0.01). CONCLUSIONS: Computer-based simulations can be a useful tool for improving ESA prescription abilities among nephrologists by raising awareness about the pharmacokinetic characteristics of the various ESAs and recognizing the factors that influence haemoglobin variability.

The use of the vac in complex wounds - does it work and how?

Background: Complex wounds pose a major challenge in reconstructive and trauma surgery. Several approaches to increase the healing process have been proposed in the last decades. In this study we study the mechanism of action of the Vacuum Assisted Closure device in diabetic wounds. Methods: Full-thickness wounds were excised in diabetic mice and treated with the VAC device or its isolated components: an occlusive dressing (OD) alone, subathmospheric pressure at 125 mm Hg (Suction), and a polyurethane foam without (Foam) and with (Foamc) downward compression of approximately 125 mm Hg. The last goups were treated with either the complete VAC device (VAC) or with a silicne interface that alows fluid removel (Mepithel-VAC). The effects of the treatment modes on the wound surface were quantified by a two-dimensional immunohistochemical staging system based on vasculature, as defined by blood vessel density (CD31) and cell proliferation (defined by ki67 positivity), 7 days post wounding. Finite element modelling was used to predict wound surface deformation under dressing modes and cross sections of in situ fixed tissues were used to measure actual microstrain. Results: The foam-wound interface of the Vacuum Assisted Closure device causes significant wound stains (60%) causing a deformation of the single cell level leading to a profound upregulation of cell proliferation (4-fold) and angiogenisis (2.2-fold) compared to OD treated wounds. Polyurethane foam exposure itself causes a frather unspecific angiogenic response (Foamc, 2 - fold, Foam, 2.2 - fold) without changes of the cell proliferation rate of the wound bed. Suction alone without a specific interface does not have an effect on meassured parameters, showing similar results to untreated wounds. A perforated silicone interface caused a significant lower microdeforamtion of the wound bed correlating to changes of the wound tissues. Conclusion: The Vacuum Assisted Closure device induce significanttissue growth in diabetic wounds. The wound foam interface under suction causes profound macrodeformation that stimulates tissue growth by angiogenesis and cell proliferation. It needs to be taken in consideration that in the clinical setting different wound types may profit from different elements of this suction device.