Do pseudo-absence selection strategies influence species distribution models and their predictions? An information-theoretic approach based on simulated data

Background Multiple logistic regression is precluded from many practical applications in ecology that aim to predict the geographic distributions of species because it requires absence data, which are rarely available or are unreliable. In order to use multiple logistic regression, many studies have simulated "pseudo-absences" through a number of strategies, but it is unknown how the choice of strategy influences models and their geographic predictions of species. In this paper we evaluate the effect of several prevailing pseudo-absence strategies on the predictions of the geographic distribution of a virtual species whose "true" distribution and relationship to three environmental predictors was predefined. We evaluated the effect of using a) real absences b) pseudo-absences selected randomly from the background and c) two-step approaches: pseudo-absences selected from low suitability areas predicted by either Ecological Niche Factor Analysis: (ENFA) or BIOCLIM. We compared how the choice of pseudo-absence strategy affected model fit, predictive power, and information-theoretic model selection results. Results Models built with true absences had the best predictive power, best discriminatory power, and the "true" model (the one that contained the correct predictors) was supported by the data according to AIC, as expected. Models based on random pseudo-absences had among the lowest fit, but yielded the second highest AUC value (0.97), and the "true" model was also supported by the data. Models based on two-step approaches had intermediate fit, the lowest predictive power, and the "true" model was not supported by the data. Conclusion If ecologists wish to build parsimonious GLM models that will allow them to make robust predictions, a reasonable approach is to use a large number of randomly selected pseudo-absences, and perform model selection based on an information theoretic approach. However, the resulting models can be expected to have limited fit.

A new neuropsychological instrument measuring effects of age and drugs on fitness to drive: development, reliability, and validity of MedDrive

Background: Current guidelines underline the limitations of existing instruments to assess fitness to drive and the poor adaptability of batteries of neuropsychological tests in primary care settings. Aims: To provide a free, reliable, transparent computer based instrument capable of detecting effects of age or drugs on visual processing and cognitive functions. Methods: Relying on systematic reviews of neuropsychological tests and driving performances, we conceived four new computed tasks measuring: visual processing (Task1), movement attention shift (Task2), executive response, alerting and orientation gain (Task3), and spatial memory (Task4). We then planned five studies to test MedDrive's reliability and validity. Study-1 defined instructions and learning functions collecting data from 105 senior drivers attending an automobile club course. Study-2 assessed concurrent validity for detecting minor cognitive impairment (MCI) against useful field of view (UFOV) on 120 new senior drivers. Study-3 collected data from 200 healthy drivers aged 20-90 to model age related normal cognitive decline. Study-4 measured MedDrive's reliability having 21 healthy volunteers repeat tests five times. Study-5 tested MedDrive's responsiveness to alcohol in a randomised, double-blinded, placebo, crossover, dose-response validation trial including 20 young healthy volunteers. Results: Instructions were well understood and accepted by all senior drivers. Measures of visual processing (Task1) showed better performances than the UFOV in detecting MCI (ROC 0.770 vs. 0.620; p=0.048). MedDrive was capable of explaining 43.4% of changes occurring with natural cognitive decline. In young healthy drivers, learning effects became negligible from the third session onwards for all tasks except for dual tasking (ICC=0.769). All measures except alerting and orientation gain were affected by blood alcohol concentrations. Finally, MedDrive was able to explain 29.3% of potential causes of swerving on the driving simulator. Discussion and conclusions: MedDrive reveals improved performances compared to existing computed neuropsychological tasks. It shows promising results both for clinical and research purposes.

Land use improves spatial predictions of mountain plant abundance but not presence-absence

Question Does a land-use variable improve spatial predictions of plant species presence-absence and abundance models at the regional scale in a mountain landscape? Location Western Swiss Alps. Methods Presence-absence generalized linear models (GLM) and abundance ordinal logistic regression models (LRM) were fitted to data on 78 mountain plant species, with topo-climatic and/or land-use variables available at a 25-m resolution. The additional contribution of land use when added to topo-climatic models was evaluated by: (1) assessing the changes in model fit and (2) predictive power, (3) partitioning the deviance respectively explained by the topo-climatic variables and the land-use variable through variation partitioning, and (5) comparing spatial projections. Results Land use significantly improved the fit of presence-absence models but not their predictive power. In contrast, land use significantly improved both the fit and predictive power of abundance models. Variation partitioning also showed that the individual contribution of land use to the deviance explained by presence-absence models was, on average, weak for both GLM and LRM (3.7% and 4.5%, respectively), but changes in spatial projections could nevertheless be important for some species. Conclusions In this mountain area and at our regional scale, land use is important for predicting abundance, but not presence-absence. The importance of adding land-use information depends on the species considered. Even without a marked effect on model fit and predictive performance, adding land use can affect spatial projections of both presence-absence and abundance models.

PAR2 absence completely rescues inflammation and ichthyosis caused by altered CAP1/Prss8 expression in mouse skin.

Altered serine protease activity is associated with skin disorders in humans and in mice. The serine protease channel-activating protease-1 (CAP1; also termed protease serine S1 family member 8 (Prss8)) is important for epidermal homeostasis and is thus indispensable for postnatal survival in mice, but its roles and effectors in skin pathology are poorly defined. In this paper, we report that transgenic expression in mouse skin of either CAP1/Prss8 (K14-CAP1/Prss8) or protease-activated receptor-2 (PAR2; Grhl3(PAR2/+)), one candidate downstream target, causes epidermal hyperplasia, ichthyosis and itching. K14-CAP1/Prss8 ectopic expression impairs epidermal barrier function and causes skin inflammation characterized by an increase in thymic stromal lymphopoietin levels and immune cell infiltrations. Strikingly, both gross and functional K14-CAP1/Prss8-induced phenotypes are completely negated when superimposed on a PAR2-null background, establishing PAR2 as a pivotal mediator of pathogenesis. Our data provide genetic evidence for PAR2 as a downstream effector of CAP1/Prss8 in a signalling cascade that may provide novel therapeutic targets for ichthyoses, pruritus and inflammatory skin diseases.

Increased epidermal expression and absence of mutations in CARD14 in a series of patients with sporadic pityriasis rubra pilaris.

Pityriasis rubra pilaris (PRP; MIM 173200) encompasses a spectrum of rare chronic papulosquamous inflammatory disorders, which have been classified into 6 subtypes(1) . Clinical features include palmoplantar keratoderma and follicular hyperkeratotic papules which coalesce into large, scaly, erythematous plaques, with frequent progression to exfoliative erythroderma. This article is protected by copyright. All rights reserved.

Absence of the fourth cranial nerve in congenital Brown syndrome.

PURPOSE: To elucidate the aetiology of congenital Brown syndrome. METHODS: Four consecutive patients diagnosed with unilateral congenital Brown syndrome had a comprehensive standardized ocular motility examination. Any compensatory head posture was measured. Brain magnetic resonance imaging (MRI) with regard for the IV cranial nerve (CN) was performed in all patients. Orbital MRI was performed in 2/4 patients, with images acquired in eight directions of gaze and superior oblique (SO) muscle areas compared. RESULTS: CN IV could not be identified bilaterally in two patients, but was absent only on the side of the Brown syndrome in the two other patients. On the normal side, orbital MRI revealed a smaller SO muscle area in upgaze than in downgaze, demonstrating normal actions of this muscle. On the side of the Brown syndrome, the SO area remained the same in upgaze and in downgaze and approximately symmetric to the area of SO in downgaze on the normal side. CONCLUSIONS: These cases add further anatomical support to the theory of paradoxical innervation in congenital Brown syndrome. CN IV was absent in two patients on the side of the Brown syndrome, but without muscle hypoplasia. SO muscle size did not vary in up- and downgaze, which we interpreted as a sign of constant innervation through branches of CN III.

Discriminant validity and test-retest reliability of a self-administered Internet-based questionnaire testing doctors' knowledge in evidence-based medicine.

Aims and objectives  This study aimed to determine the discriminant validity and the test-retest reliability of a questionnaire testing the impact of evidence-based medicine (EBM) training on doctors' knowledge and skills. Methods  Questionnaires were sent electronically to all doctors working as residents and chief residents in two French speaking hospital networks in Switzerland. Participants completed the questionnaire twice, within a 4-week interval. The discriminant validity was examined in comparing doctors' performance according to their reported EBM previous training. Proportion of agreement between both sessions of the questionnaire, Cohen's kappa and 'uniform kappa' determined its test-retest reliability. Results  The participation rate was 9.8%/7.1% to first/second session. Performance increased according to the level of doctors' previous training in EBM. The observed proportion of agreement between both sessions was over 70% for 14/19 questions, and the 'uniform kappa' was superior to 0.60 for 15/19 questions. Conclusion  The discriminant validity and test-retest reliability of the questionnaire were satisfying. The low participation rate did not prevent the study from achieving its aims.

Shoulder sensorimotor control assessment by force platform: feasibility and reliability.

Given the important role of the shoulder sensorimotor system in shoulder stability, its assessment appears of interest. Force platform monitoring of centre of pressure (CoP) in upper-limb weight-bearing positions is of interest as it allows integration of all aspects of shoulder sensorimotor control. This study aimed to determine the feasibility and reliability of shoulder sensorimotor control assessment by force platform. Forty-five healthy subjects performed two sessions of CoP measurement using Win-Posturo(®) Medicapteurs force platform in an upper-limb weight-bearing position with the lower limbs resting on a table to either the anterior superior iliac spines (P1) or upper patellar poles (P2). Four different conditions were tested in each position in random order: eyes open or eyes closed with trunk supported by both hands and eyes open with trunk supported on the dominant or non-dominant side. P1 reliability values were globally moderate to high for CoP length, CoP velocity and CoP standard deviation (SD), standard error of measurement ranged from 6·0% to 26·5%, except for CoP area. P2 reliability values were globally low and not clinically acceptable. Our results suggest that shoulder sensorimotor control assessment by force platform is feasible and has good reliability in upper-limb weight-bearing positions when the lower limbs are resting on a table to the anterior superior iliac spines. CoP length, CoP velocity and CoP SD velocity appear to be the most reliable variables.

Transcutaneous carbon dioxide tension in newborn infants: reliability and safety of continuous 24-hour measurement at 42 degrees C.

In 58 newborn infants a new iridium oxide sensor was evaluated for transcutaneous carbon dioxide (tcPCO2) monitoring at 42 degrees C with a prolonged fixation time of 24 hours. The correlation of tcPCO2 (y; mm Hg) v PaCO2 (x; mm Hg) for 586 paired values was: y = 4.6 + 1.45x; r = .89; syx = 6.1 mm Hg. The correlation was not influenced by the duration of fixation. The transcutaneous sensor detected hypocapnia (PaCO2 less than 35 mm Hg) in 74% and hypercapnia (PCO2 greater than 45 mm Hg) in 74% of all cases. After 24 hours, calibration shifts were less than 4 mm Hg in 90% of the measuring periods. In 86% of the infants, no skin changes were observed; in 12% of infants, there were transitional skin erythemas and in 2% a blister which disappeared without scarring. In newborn infants with normal BPs, continuous tcPCO2 monitoring at 42 degrees C can be extended for as many as 24 hours without loss of reliability or increased risk for skin burns.

Validation study of the French version of the observer-rater scales measuring coping and cognitive errors (CE-CAP) on a non-clinical population

This study examined the validity and reliability of the French version of two observer-rated measures developed to assess cognitive errors (cognitive errors rating system [CERS]) [6] and coping action patterns (coping action patterns rating system [CAPRS]) [22,24]. The CE measures 14 cognitive errors, broken down according to their valence positive or negative (see the definitions by A.T. Beck), and the CAP measures 12 coping categories, based on an comprehensive review literature, each broken down into three levels of action (affective, behavioural, cognitive). Thirty (N = 30) subjects recruited in a community sample participated in the study. They were interviewed according to a standardized clinical protocol: these interviews were transcribed and analysed with both observer-rated systems. Results showed that the inter-rater reliability of the two measures is good and that their internal validity is satisfactory, due to a non-significant canonical correlation between CAP and CE. With regard to discriminant validity, we found a non-significant canonical correlation between CAPRS and CISS, one of most widely used self-report questionnaire measuring coping. The same can be said for the correlation with a self-report questionnaire measuring symptoms (SCL-90-R). These results confirm the absence of confounds in the assessment of cognitive errors and of coping as assessed by these observer-rated scales and add an argument in favour of the French validation of the CE-CAP rating scales. (C) 2010 Elsevier Masson SAS. All rights reserved.

"INTERMED": a method to assess health service needs. I. Development and reliability.

The purpose of this paper is to describe the development and to test the reliability of a new method called INTERMED, for health service needs assessment. The INTERMED integrates the biopsychosocial aspects of disease and the relationship between patient and health care system in a comprehensive scheme and reflects an operationalized conceptual approach to case mix or case complexity. The method is developed to enhance interdisciplinary communication between (para-) medical specialists and to provide a method to describe case complexity for clinical, scientific, and educational purposes. First, a feasibility study (N = 21 patients) was conducted which included double scoring and discussion of the results. This led to a version of the instrument on which two interrater reliability studies were performed. In study 1, the INTERMED was double scored for 14 patients admitted to an internal ward by a psychiatrist and an internist on the basis of a joint interview conducted by both. In study 2, on the basis of medical charts, two clinicians separately double scored the INTERMED in 16 patients referred to the outpatient psychiatric consultation service. Averaged over both studies, in 94.2% of all ratings there was no important difference between the raters (more than 1 point difference). As a research interview, it takes about 20 minutes; as part of the whole process of history taking it takes about 15 minutes. In both studies, improvements were suggested by the results. Analyses of study 1 revealed that on most items there was considerable agreement; some items were improved. Also, the reference point for the prognoses was changed so that it reflected both short- and long-term prognoses. Analyses of study 2 showed that in this setting, less agreement between the raters was obtained due to the fact that the raters were less experienced and the scoring procedure was more susceptible to differences. Some improvements--mainly of the anchor points--were specified which may further enhance interrater reliability. The INTERMED proves to be a reliable method for classifying patients' care needs, especially when used by experienced raters scoring by patient interview. It can be a useful tool in assessing patients' care needs, as well as the level of needed adjustment between general and mental health service delivery. The INTERMED is easily applicable in the clinical setting at low time-costs.

Evidence for validity and reliability of a french version of the FAAM.

BACKGROUND: The Foot and Ankle Ability Measure (FAAM) is a self reported questionnaire for patients with foot and ankle disorders available in English, German, and Persian. This study plans to translate the FAAM from English to French (FAAM-F) and assess the validity and reliability of this new version.METHODS: The FAAM-F Activities of Daily Living (ADL) and sports subscales were completed by 105 French-speaking patients (average age 50.5 years) presenting various chronic foot and ankle disorders. Convergent and divergent validity was assessed by Pearson's correlation coefficients between the FAAM-F subscales and the SF-36 scales: Physical Functioning (PF), Physical Component Summary (PCS), Mental Health (MH) and Mental Component Summary (MCS). Internal consistency was calculated by Cronbach's Alpha (CA). To assess test re-test reliability, 22 patients filled out the questionnaire a second time to estimate minimal detectable changes (MDC) and intraclass correlation coefficients (ICC).RESULTS: Correlations for FAAM-F ADL subscale were 0.85 with PF, 0.81 with PCS, 0.26 with MH, 0.37 with MCS. Correlations for FAAM-F Sports subscale were 0.72 with PF, 0.72 with PCS, 0.21 with MH, 0.29 with MCS. CA estimates were 0.97 for both subscales. Respectively for the ADL and Sports subscales, ICC were 0.97 and 0.94, errors for a single measure were 8 and 10 points at 95% confidence and the MDC values at 95% confidence were 7 and 18 points.CONCLUSION: The FAAM-F is valid and reliable for the self-assessment of physical function in French-speaking patients with a wide range of chronic foot and ankle disorders.