Multivariable adjustments counteract spectrum and test review bias in accuracy studies

OBJECTIVES: The STAndards for Reporting studies of Diagnostic accuracy (STARD) for investigators and editors and the Quality Assessment of Diagnostic Accuracy Studies (QUADAS) for reviewers and readers offer guidelines for the quality and reporting of test accuracy studies. These guidelines address and propose some solutions to two major threats to validity: spectrum bias and test review bias. STUDY DESIGN AND SETTING: Using a clinical example, we demonstrate that these solutions fail and propose an alternative solution that concomitantly addresses both sources of bias. We also derive formulas that prove the generality of our arguments. RESULTS: A logical extension of our ideas is to extend STARD item 23 by adding a requirement for multivariable statistical adjustment using information collected in QUADAS items 1, 2, and 12 and STARD items 3-5, 11, 15, and 18. CONCLUSION: We recommend reporting not only variation of diagnostic accuracy across subgroups (STARD item 23) but also the effects of the multivariable adjustments on test performance. We also suggest that the QUADAS be supplemented by an item addressing the appropriateness of statistical methods, in particular whether multivariable adjustments have been included in the analysis.

Bovine Tuberculosis Risk Factors for British Herds Before and After the 2001 Foot-and-Mouth Epidemic: What have we Learned from the TB99 and CCS2005 Studies?

Over the last couple of decades, the UK experienced a substantial increase in the incidence and geographical spread of bovine tuberculosis (TB), in particular since the epidemic of foot-and-mouth disease (FMD) in 2001. The initiation of the Randomized Badger Culling Trial (RBCT) in 1998 in south-west England provided an opportunity for an in-depth collection of questionnaire data (covering farming practices, herd management and husbandry, trading and wildlife activity) from herds having experienced a TB breakdown between 1998 and early 2006 and randomly selected control herds, both within and outside the RBCT (the so-called TB99 and CCS2005 case-control studies). The data collated were split into four separate and comparable substudies related to either the pre-FMD or post-FMD period, which are brought together and discussed here for the first time. The findings suggest that the risk factors associated with TB breakdowns may have changed. Higher Mycobacterium bovis prevalence in badgers following the FMD epidemic may have contributed to the identification of the presence of badgers on a farm as a prominent TB risk factor only post-FMD. The strong emergence of contact/trading TB risk factors post-FMD suggests that the purchasing and movement of cattle, which took place to restock FMD-affected areas after 2001, may have exacerbated the TB problem. Post-FMD analyses also highlighted the potential impact of environmental factors on TB risk. Although no unique and universal solution exists to reduce the transmission of TB to and among British cattle, there is an evidence to suggest that applying the broad principles of biosecurity on farms reduces the risk of infection. However, with trading remaining as an important route of local and long-distance TB transmission, improvements in the detection of infected animals during pre- and post-movement testing should further reduce the geographical spread of the disease.

Ultrafast Relaxation Dynamics of Osmium-Polypyridine Complexes in Solution

We present steady-state absorption and emission spectroscopy and femtosecond broadband photoluminescence up-conversion spectroscopy studies of the electronic relaxation of Os(dmbp)3 (Os1) and Os(bpy)2(dpp) (Os2) in ethanol, where dmbp is 4,4′-dimethyl-2,2′-biypridine, bpy is 2,2′-biypridine, and dpp is 2,3-dipyridyl pyrazine. In both cases, the steady-state phosphorescence is due to the lowest 3MLCT state, whose quantum yield we estimate to be ≤5.0 × 10–3. For Os1, the steady-state phosphorescence lifetime is 25 ns. In both complexes, the photoluminescence excitation spectra map the absorption spectrum, pointing to an excitation wavelength-independent quantum yield. The ultrafast studies revealed a short-lived (≤100 fs) fluorescence, which stems from the lowest singlet metal-to-ligand-charge-transfer (1MLCT) state and decays by intersystem crossing to the manifold of 3MLCT states. In addition, Os1 exhibits a 50 ps lived emission from an intermediate triplet state at an energy 2000 cm–1 above that of the long-lived (25 ns) phosphorescence. In Os2, the 1MLCT–3MLCT intersystem crossing is faster than that in Os1, and no emission from triplet states is observed other than the lowest one. These observations are attributed to a higher density of states or a smaller energy spacing between them compared with Os1. They highlight the importance of the energetics on the rate of intersystem crossing.

Water–montmorillonite systems: Neutron scattering and tracer throughdiffusion studies

Designs for deep geological respositories of nuclear waste include bentonite as a hydraulic and chemisorption buffer material to protect the biosphere from leakage of radionuclides. Bentonite is chosen because it is a cheap, naturally occurring material with the required properties. It consists essentially of montmorillonite, a swelling clay mineral. Upon contact with groundwater such clays can seal the repository by incorporating water in the interlayers of their crystalline structure. The intercalated water exhibits significantly different properties to bulk water in the surrounding interparticle pores, such as lower diffusion coefficients (González Sánchez et. al. 2008). This doctoral thesis presents water distribution and diffusion behavior on various time and space scales in montmorillonite. Experimental results are presented for Na- and Cs-montmorillonite samples with a range of bulk dry densities (0.8 to 1.7 g/cm3). The experimental methods employed were neutron scattering (backscattering, diffraction, time-of-flight), adsorption measurements (water, nitrogen) and tracer-through diffusion. For the tracer experiments the samples were fully saturated via the liquid phase under volume-constrained conditions. In contrast, for the neutron scattering experiments, the samples were hydrated via the vapor phase and subsequently compacted, leaving a significant fraction of interparticle pores unfilled with water. Owing to these differences in saturation, the water contents of the samples for neutron scattering were characterized by gravimetry whereas those for the tracer experiments were obtained from the bulk dry density. The amount of surface water in interlayer pores could be successfully discriminated from the amount of bulk-like water in interparticle pores in Na- and Csmontmorillonite using neutron spectroscopy. For the first time in the literature, the distribution of water between these two pore environments was deciphered as a function of gravimetric water content. The amount was compared to a geometrical estimation of the amount of interlayer and interparticle water determined by neutron diffraction and adsorption measurements. The relative abundances of the 1 to 4 molecular water layers in the interlayer were determined from the area ratios of the (001)-diffraction peaks. Depending on the characterization method, different fractions of surface water and interlayer water were obtained. Only surface and interlayer water exists in amontmorillonite with water contents up to 0.18 g/g according to spectroscopic measurements and up to 0.32 g/g according to geometrical estimations, respectively. At higher water contents, bulk-like and interparticle water also exists. The amounts increase monotonically, but not linearly, from zero to 0.33 g/g for bulk-like water and to 0.43 g/g for interparticle water. It was found that water most likely redistributes between the surface and interlayer sites during the spectroscopic measurements and therefore the reported fraction is relevant only below about -10 ºC (Anderson, 1967). The redistribution effect can explain the discrepancy in fractions between the methods. In a novel approach the fractions of water in different pore environments were treated as a fixed parameter to derive local diffusion coefficients for water from quasielastic neutron scattering data, in particular for samples with high water contents. Local diffusion coefficients were obtained for the 1 to 4 molecular water layers in the interlayer of 0.5·10–9, 0.9·10–9, 1.5·10–9 and 1.4·10–9 m²/s, respectively, taking account of the different water fractions (molecular water layer, bulk-like water). The diffusive transport of 22Na and HTO through Na-montmorillonite was measured on the laboratory experimental scale (i.e. cm, days) by tracer through-diffusion experiments. We confirmed that diffusion of HTO is independent of the ionic strength of the external solution in contact with the clay sample but dependent on the bulk dry density. In contrast, the diffusion of 22Na was found to depend on both the ionic strength of the pore solution and on the bulk dry density. The ratio of the pore and surface diffusion could be experimentally determined for 22Na from the dependence of the diffusion coefficient on the ionic strength. Activation energies were derived from the temperaturedependent diffusion coefficients via the Arrhenius relation. In samples with high bulk dry density the activation energies are slightly higher than those of bulk water whereas in low density samples they are lower. The activation energies as a function of ionic strengths of the pore solutions are similar for 22Na and HTO. The facts that (i) the slope of the logarithmic effective diffusion coefficients as a function of the logarithmic ionic strength is less than unity for low bulk dry densities and (ii) two water populations can be observed for high gravimetric water contents (low bulk dry densities) support the interlayer and interparticle porosity model proposed by Glaus et al. (2007), Bourg et al. (2006, 2007) and Gimmi and Kosakowski (2011).