Uncertainty In Model Results Research Articles

Cost-Effectiveness of Nadofaragene Firadenovec and Pembrolizumab in Bacillus Calmette-Guérin Immunotherapy Unresponsive Non–Muscle Invasive Bladder Cancer

Nadofaragene firadenovec is a gene therapy for bacillus Calmette-Guérin (BCG)-unresponsive non-muscle-invasive bladder cancer (NMIBC) undergoing Food and Drug Administration review. Pembrolizumab is approved for treating patients with BCG-unresponsive NMIBC with carcinoma in situ (CIS). We evaluated the cost-effectiveness of these treatments compared with a hypothetical therapeutic alternative, at a willingness-to-pay threshold of $150 000 per quality-adjusted life-year (QALY) gained, in CIS and non-CIS BCG-unresponsive NMIBC populations. We developed a Markov cohort simulation model with a 3-month cycle length and lifetime horizon to estimate the total costs, QALYs, and cost per additional QALY from the health sector perspective. Clinical inputs were informed by results of single-arm clinical trials evaluating the treatments, and systematic literature reviews were conducted to obtain other model inputs. Sensitivity analyses were conducted to assess uncertainty in model results. Nadofaragene firadenovec, at a placeholder price 10% higher than the price of pembrolizumab, had an incremental cost-effectiveness ratio of $263 000 and $145 000 per QALY gained in CIS and non-CIS populations, respectively. Pembrolizumab had an incremental cost-effectiveness ratio of $168 000 per QALY gained for CIS. A 5.4% reduction in pembrolizumab's price would make it cost-effective. The model was sensitive to many inputs, especially to the probabilities of disease progression, initial treatment response and durability, and drug price. The cost-effectiveness of nadofaragene firadenovec will depend upon its price. Pembrolizumab, although not cost-effective in our base-case analysis, is an important alternative in this population with an unmet medical need. Comparative trials of these treatments are warranted to better estimate cost-effectiveness.

Prediction scoring of data-driven discoveries for reproducible research

Predictive modeling uncovers knowledge and insights regarding a hypothesized data generating mechanism (DGM). Results from different studies on a complex DGM, derived from different data sets, and using complicated models and algorithms, are hard to quantitatively compare due to random noise and statistical uncertainty in model results. This has been one of the main contributors to the replication crisis in the behavioral sciences. The contribution of this paper is to apply prediction scoring to the problem of comparing two studies, such as can arise when evaluating replications or competing evidence. We examine the role of predictive models in quantitatively assessing agreement between two datasets that are assumed to come from two distinct DGMs. We formalize a distance between the DGMs that is estimated using cross validation. We argue that the resulting prediction scores depend on the predictive models created by cross validation. In this sense, the prediction scores measure the distance between DGMs, along the dimension of the particular predictive model. Using human behavior data from experimental economics, we demonstrate that prediction scores can be used to evaluate preregistered hypotheses and provide insights comparing data from different populations and settings. We examine the asymptotic behavior of the prediction scores using simulated experimental data and demonstrate that leveraging competing predictive models can reveal important differences between underlying DGMs. Our proposed cross-validated prediction scores are capable of quantifying differences between unobserved data generating mechanisms and allow for the validation and assessment of results from complex models.

Using the finite element method to evaluate the load-bearing capacity of beams with a corrugated web subjected to local loading

Introduction. Steel beams with a corrugated web are increasingly frequently applied in industrial and civil engineering due to their cost effectiveness. There are numerous studies proving the advantages of corrugated beams. However, the issue of calculating such elements is insufficiently covered in the standards and engineering literature, which is one of the main factors restraining their widespread use.
 
 Materials and methods. Along with the use of analytical dependencies, which, as a rule, are focused on a specific type of web corrugation and have limitations in terms of the area of experimental susceptibility, numerical methods are widely used. The finite element (FE) method is applied in the article.
 
 Results. The article presents the principles of constructing a FE model for evaluating the load-bearing capacity and service ability of corrugated beams subjected to local loading (patch loading), verified by using the experimental data. The authors have analysed the influence of parameters of the FE model and input variables on the accuracy and uncertainty of modeling results.
 
 Conclusions. The article shows that the use of FE models allows for a highly accurate evaluation of the load-bearing capacity and the behaviour of a beam with a corrugated web subjected to local loading. The description of the behaviour of steel has one of dominant influences on the accuracy of FE models, while the value of yield strength has a dominant influence; values of ultimate strength and the type of deformation diagram have an auxiliary influence. The variability in the web thickness has a directly proportional effect on the value of the bearing capacity. The size of the finite element should be determined according to the condition of convergence of results against the criteria of critical and ultimate forces. The most optimal size of the finite element is about 3–5 web thicknesses. To reduce the total number of finite elements, it is recommended to use local condensation in areas of stresses. The shape of equivalent geometric imperfections is recommended to be assigned based on the forms of elastic buckling of the web.

Thermal state of the lithosphere in Eastern Paraguay and in Andean Domain (South American Platform)

Crustal thermal models that incorporate thermo-barometric data have been developed for estimating depth to 1300 ºC isotherm in two xenoliths provinces: Southeast Paraguay and Andean domain, in South American Platform. Uncertainties in model results has been minimized by imposing reasonable bounds on some of the key model parameters. Considering only the best fit results it is possible to infer average values for geothermal parameters at the surface. This imply heat flow of 86 mWm-2, radiogenic heat production of 1.8 µWm-3. Besides at Moho depth: heat flow of 21 mWm-2, radiogenic heat production of 4.5x10-3 µWm-3, temperature of from Southeast Paraguay. For the Andean Domain, we have the following values for the geothermal parameters: heat flow, 72 mWm-2, radiogenic heat production, 1.0 µWm-3 in surface and heat flow of 33 mWm-2, radiogenic heat production of 2.0x10-3 µWm-3 and temperature of 785ºC in Moho depth. The heat flux estimated for the southeastern Paraguay is higher than that for the Andean domain. This result is in agreement with differences in geological ages between these sites, since the age value for Paraguayan region is approximately 20% lower than the Andean one.

Estimating Permeability of Partially Frozen Soil Using Floating Random Walks

AbstractFlow through partially frozen pores in granular media containing ice or gas hydrate plays an essential role in diverse phenomena including methane migration and frost heave. As freezing progresses, the frozen phase grows in the pore space and constricts flow paths so that the permeability decreases. Previous works have measured the relationship between permeability and volumetric fraction of the frozen phase, and various correlations have been proposed to predict permeability change in hydrology and the oil industry. However, predictions from different formulae can differ by orders of magnitude, causing great uncertainty in modeling results. We present a floating random walk method to approximate the porous flow field and estimate the effective permeability in isotropic granular media with specified particle size distributions, without solving for the entire flow field in the pore space. In packed spherical particles, the method compares favorably with the Kozeny‐Carman formula. We further extend this method with a probabilistic interpretation of the volumetric fraction of the frozen phase, simulate the effect of freezing in irregular pores, and predict the evolution of permeability. Employing no adjustable parameters, our results can provide insight into the coupling between phase transitions and permeability change, which plays important roles in hydrate formation and dissociation, as well as in the thawing and freezing of permafrost and ice‐bed coupling beneath glaciers.

Assessing uncertainties from physical parameters and modelling choices in an atmospheric large eddy simulation model.

In this study, we investigate uncertainties in a large eddy simulation of the atmosphere, employing modern uncertainty quantification methods that have hardly been used yet in this context. When analysing the uncertainty of model results, one can distinguish between uncertainty related to physical parameters whose values are not exactly known, and uncertainty related to modelling choices such as the selection of numerical discretization methods, of the spatial domain size and resolution, and the use of different model formulations. While the former kind is commonly studied e.g. with forward uncertainty propagation, we explore the use of such techniques to also assess the latter kind. From a climate modelling perspective, uncertainties in the convective response and cloud formation are of particular interest, since these affect the cloud-climate feedback, one of the dominant sources of uncertainty in current climate models. Therefore we analyse the DALES model in the RICO case, a well-studied convection benchmark. We use the VECMA toolkit for uncertainty propagation, assessing uncertainties stemming from physical parameters as well as from modelling choices. We find substantial uncertainties due to small random initial state perturbations, and that the choice of advection scheme is the most influential of the modelling choices we assessed.This article is part of the theme issue ‘Reliability and reproducibility in computational science: implementing verification, validation and uncertainty quantification in silico’.

Nonlinear Double-Log Mixing Rule for Viscosity Calculation of Bitumen/Solvent Mixtures Applicable for Reservoir Simulation of Solvent-Based Recovery Processes

SummaryVarious solvent-based recovery processes for bitumen and heavy-oil reservoirs have gained much interest in recent years. In these processes, viscosity reduction is attained not only because of thermal effects, but also by dilution of bitumen with a solvent. Accurate characterization of the oil/solvent-mixture viscosity is critical for accurate prediction of recovery and effectiveness of such processes. There are varieties of models designed to predict and correlate the mixture viscosities. Among them, the linear log mixing (Arrhenius) model is the most commonly used method in the oil industry. This model, originally proposed for light oils, often show poor performance (40 to 60% error) when applied to highly viscous fluids such as heavy oil and bitumen. The modified Arrhenius model, called the nonlinear log mixing model, gives slightly better predictions compared with the original Arrhenius model. However, the predictions still might not be acceptable because of large deviations from measured experimental data. Calculated mixture-phase viscosity has a significant effect on flow calculations in commercial reservoir simulators. Underestimation of mixture viscosities leads to overprediction of oil-production rates. Using such mixing models in reservoir simulation can lead to inaccuracy in mixture viscosities and hence large uncertainty in model results. In the present study, different correlations and mixing rules available in the literature are evaluated against the mixture-viscosity data for a variety of bitumen/solvent systems. A new form (nonlinear) of the double-log mixing rule is proposed, which shows a significant improvement over the existing models on predicting viscosities of bitumen/solvent mixtures, especially at high temperatures.

Influence of watershed characteristics on streambed hydraulic conductivity across multiple stream orders

Streambeds are critical hydrological interfaces: their physical properties regulate the rate, timing, and location of fluxes between aquifers and streams. Streambed vertical hydraulic conductivity (Kv) is a key parameter in watershed models, so understanding its spatial variability and uncertainty is essential to accurately predicting how stresses and environmental signals propagate through the hydrologic system. Most distributed modeling studies use generalized Kv estimates from column experiments or grain-size distribution, but Kv may include a wide range of orders of magnitude for a given particle size group. Thus, precisely predicting Kv spatially has remained conceptual, experimental, and/or poorly constrained. This usually leads to increased uncertainty in modeling results. There is a need to shift focus from scaling up pore-scale column experiments to watershed dimensions by proposing a new kind of approach that can apply to a whole watershed while incorporating spatial variability of complex hydrological processes. Here we present a new approach, Multi-Stemmed Nested Funnel (MSNF), to develop pedo-transfer functions (PTFs) capable of simulating the effects of complex sediment routing on Kv variability across multiple stream orders in Frenchman Creek watershed, USA. We find that using the product of Kv and drainage area as a response variable reduces the fuzziness in selecting the “best” PTF. We propose that the PTF can be used in predicting the ranges of Kv values across multiple stream orders.

An overview of optimization technologies applied in combined cooling, heating and power systems

Combined cooling, heating and power (CCHP) systems have the advantages of higher energy efficiency and lower economic and environmental impacts (3E) than separated systems. Recently, optimization methods have been extensively applied in CCHP system's design and control for further improving 3E performances of CCHP systems. However, a wide range of subsystems selection, nonlinear characteristics of subsystem models and uncertainty in modeling results of CCHP systems cause challenges for successful implementation of optimization studies at both design and control stage. This paper presents a systematic review of all aspects of optimization for CCHP systems from problem formulation and algorithms selection to technical implementation. Genetic algorithm, particle swarm optimization and differential evolution are the most common algorithms used in system optimum design and control. Optimization works in design stage focus on optimization of energy source, prime mover, storage system, energy demand and system configuration. Over 60% of renewable energy integrated CCHP system adopted solar energy. More investigations are needed for small scale prime mover and economic is the development obstacle among 3E objectives. Dynamic characteristics and mal-distribution problem should earn more attention when optimizing storage system in CCHP. For the system control aspects, the baseline control strategy of load-following method as well as optimum control under real-time data and uncertainty condition are discussed. The intermittent feature of renewable energy should be considered at system control stage with uncertainty study. Uncertainty inside of energy demands, subsystems performance and market price should be investigated at the same time during control stage. Overall, this review paper can be used as an optimization reference and guidance for optimum CCHP system design and control.

OP34 One-Way Sensitivity Analysis For Cost Effectiveness Analysis

IntroductionAlthough stochastic analysis has become the accepted standard for decision analytic cost effectiveness models, deterministic one-way sensitivity analysis continues to be used to meet the needs of decision makers to understand the impact that changing the value taken by one specific parameter has on the results of the analysis. However, there are a number of problems with this approach.MethodsWe review the reasons why deterministic one-way sensitivity analysis will provide decision makers with biased and incomplete information. We then describe a new method - stochastic one-way sensitivity analysis (SOWSA), and apply this to a previously published cost effectiveness analysis, to produce a stochastic tornado diagram and conditional incremental net benefit curve. We then discuss how these outputs should be interpreted and the potential barriers to the implementation of SOWSA.ResultsThe results illustrate the shortcomings of the current approaches to deterministic one-way sensitivity analysis. For SOWSA, the expected costs and outcomes are captured, along with the sampled value of the parameter and these are linked to the probability that the parameter takes that value – which can be read off the probability distribution for the parameter used in the stochastic analysis. From these results it is possible to gain insights into probability that a parameter will take a value that will change a decision.ConclusionsAlthough a well-used technique, one-way deterministic sensitivity analysis has a number of shortcomings that may contribute to incorrect conclusions being drawn about the importance of certain parameter values on model results. By providing fuller information on uncertainty in model results, it is hoped that the methods here will lead to more informed decision making. Although, as with all developments in the presentation of analytic results to decision makers, care will be required to ensure that the decision makers understand the information provided to them.

Uncertainty Quantification in Modeling Metal Alloy Solidification

Numerical simulations of metal alloy solidification are used to gain insight into physical phenomena that cannot be observed experimentally. These models produce results that are used to draw conclusions about a process or alloy and often compared to experimental results. However, uncertainty in model inputs cause uncertainty in model results, which have the potential to significantly affect conclusions drawn from their predictions. As a step toward understanding the effect of uncertain inputs on solidification modeling, uncertainty quantification (UQ) and sensitivity analysis are performed on a transient model of solidification of Al–4.5 wt % Cu in a rectangular cavity. The binary alloy considered has columnar solidification morphology, and this model solves equations for momentum, temperature, and species conservation. UQ and sensitivity analysis are performed for the degree of macrosegregation and solidification time. A Smolyak sparse grid algorithm is used to select input values to construct a polynomial response surface fit to model outputs. This polynomial is then used as a surrogate for the complete solidification model to determine the sensitivities and probability density functions (PDFs) of the model outputs. Uncertain model inputs of interest include the secondary dendrite arm spacing (SDAS), heat transfer coefficient, and material properties. The most influential input parameter for predicting the macrosegregation level is the dendrite arm spacing, which also strongly depends on the choice of permeability model. Additionally, the degree of uncertainty required to produce accurate predictions depends on the outputs of interest from the model.

A robust parameter approach for estimating CERES-Rice model parameters for the Vietnam Mekong Delta

Rice crop models, such as CERES-Rice, are useful tools to predict rice growth and understand the effects of climatic and management changes on rice yield. As these models involve complex processes, non-linearity and interdependence of parameters may result in high uncertainty in model results. In this study, we tested the CERES-Rice model in the Vietnam Mekong Delta for two rice cultivars (Jasmine 85 and VD20) in four experimental sites. We applied a two-stage calibration procedure. In the first stage, initial sets of good performing parameters were identified. In the second stage, the good performing parameter sets and their minimum-maximum ranges were refined using the Robust Parameter Estimation (ROPE) approach. We found that in the first calibration stage, the majority of the parameter sets that performed well in the calibration sites performed poorly in the validation sites. For example, the simulated rice yields were within +/−5% of the observed yields in the calibration sites. But in the validation sites, the same parameter sets resulted in the differences of −77% to 19% for Jasmine 85 and −47% to 9.5% for VD20. This is because these parameter sets had a broad range of values, which compensated each other well in calibration but were less successful to do so in validation. When the ROPE was applied, the value ranges of the parameters narrowed down and the model performance in validation improved. In general, the parameter ranges identified by ROPE are considered more robust and have higher probability to perform better. Our positive results show a good potential of the ROPE approach for calibration of the CERES-Rice model, which can also be applied with similar other crop growth models. The robust parameter value ranges derived in this study may be used as reference parameter values for future applications of the model in the region.

Cost-Effectiveness of Intra-Arterial Treatment as an Adjunct to Intravenous Tissue-Type Plasminogen Activator for Acute Ischemic Stroke.

The objective of this study was to determine the cost-effectiveness of intra-arterial treatment within the 0- to 6-hour window after intravenous tissue-type plasminogen activator within 0- to 4.5-hour compared with intravenous tissue-type plasminogen activator alone, in the US setting and from a social perspective. A decision analytic model estimated the lifetime costs and outcomes associated with the additional benefit of intra-arterial therapy compared with standard treatment with intravenous tissue-type plasminogen activator alone. Model inputs were obtained from published literature, the Multicenter Randomized Clinical Trial of Endovascular Therapy for Acute Ischemic Stroke in the Netherlands (MR CLEAN) study, and claims databases in the United States. Health outcomes were measured in quality-adjusted life years (QALYs). Treatment benefit was assessed by calculating the cost per QALY gained. One-way and probabilistic sensitivity analyses were performed to estimate the overall uncertainty of model results. The addition of intra-arterial therapy compared with standard treatment alone yielded a lifetime gain of 0.7 QALY for an additional cost of $9911, which resulted in a cost of $14 137 per QALY. Multivariable sensitivity analysis predicted cost-effectiveness (≤$50 000 per QALY) in 97.6% of simulation runs. Intra-arterial treatment after intravenous tissue-type plasminogen activator for patients with anterior circulation strokes within the 6-hour window is likely cost-effective. From a societal perspective, increased investment in access to intra-arterial treatment for acute stroke may be justified.

Model selection and multi-model inference for Bayesian surplus production models: A case study for Pacific blue and striped marlin

Stock assessment typically involves developing a set of alternative models, fitting each to the available data, and then selecting the one that gives the most accurate estimates of management quantities of interest. In this context, it is important to consider model selection uncertainty because ignoring it can lead to unreliable estimates and overconfident inferences. For this study, four Bayesian surplus production models with symmetric or asymmetric production functions and either a constant or hierarchical time-varying intrinsic growth rate (r) were developed using data for Pacific blue marlin (Makaira nigricans) and Western and Central North Pacific striped marlin (Kajikia audax). The uncertainty resulting from model selection was evaluated using Monte Carlo simulation techniques to examine the consistency of model estimates within (self-tests) and among (cross-tests) the alternative models. Specifically, these tests evaluated the performance of the deviance information criterion (DIC) and Bayesian model averaging (BMA). The results of the simulation tests suggested that mis-specification of time-varying r can lead to large estimation errors for biomass and management quantities and that DIC may not reliably identify the true data-generating model. Although BMA did not provide more accurate point estimates than just selecting the data-generating model, it did provide a more accurate characterization of uncertainty in model results. Our study shows the value of using simulations to evaluate model performance and to account for model selection uncertainty.

Impact of Data Length on the Uncertainty of Hydrological Copula Modeling

AbstractThree Archimedean copulas were employed to model annual maximum flood peak data of different lengths. Estimation methods based on ranks were employed for parameter estimation. Marginals were modeled with the generalized extreme value (GEV) distribution. Then, uncertainty in modeling results was investigated with the change in data length. The joint and conditional return periods were also analyzed with the selected copula model to see how it varied with data length. Results showed that the accuracy of modeling deteriorated with the decrease in data length and that the best-fitting copula model depended on the data length. The uncertainty of modeling results may be due to the uncertainty of the flow itself when the data length is shortened. The data length has a negative effect not only on copula modeling but may also have an adverse effect on the marginal, which is an important factor when using a copula model to do bivariate analysis.

Quantification of Key Long-term Risks at CO2 Sequestration Sites: Latest Results from US DOE's National Risk Assessment Partnership (NRAP) Project

Risk assessment for geologic CO2 storage including quantification of risks is an area of active investigation. The National Risk Assessment Partnership (NRAP) is a US-Department of Energy (US-DOE) effort focused on developing a defensible, science-based methodology and platform for quantifying risk profiles at geologic CO2 sequestration sites. NRAP has been developing a methodology that centers round development of an integrated assessment model (IAM) using system modeling approach to quantify risks and risk profiles. The IAM has been used to calculate risk profiles with a few key potential impacts due to potential CO2 and brine leakage. The simulation results are also used to determine long-term storage security relationships and compare the long-term storage effectiveness to IPCC storage permanence goal. Additionally, we also demonstrate application of IAM for uncertainty quantification in order to determine parameters to which the uncertainty in model results is most sensitive.

Quantifying the carbon uptake by vegetation for Europe on a 1 km<sup>2</sup> resolution using a remote sensing driven vegetation model

Abstract. In this study we compare monthly gross primary productivity (GPP) time series (2000–2007), computed for Europe with the Biosphere Energy Transfer Hydrology (BETHY/DLR) model with monthly data from the eddy covariance measurements network FLUXNET. BETHY/DLR with a spatial resolution of 1 km2 is designed for regional and continental applications (here Europe) and operated at the German Aerospace Center (DLR). It was adapted from the BETHY scheme to be driven by remote sensing data (leaf area index (LAI) and land cover information) and meteorology. Time series of LAI obtained from the CYCLOPES database are used to control the phenology of vegetation. Meteorological time series from the European Centre for Medium-Range Weather Forecasts (ECMWF) are used as driver. These comprise daily information on temperature, precipitation, wind speed and radiation. Additionally, static maps such as land cover, elevation, and soil type are used. To validate our model results we used eddy covariance measurements from the FLUXNET network of 74 towers across Europe. For forest sites we found that our model predicts between 20 and 40% higher annual GPP sums. In contrast, for cropland sites BETHY/DLR results show about 18% less GPP than eddy covariance measurements. For grassland sites, between 10% more and 16% less GPP was calculated with BETHY/DLR. A mean total carbon uptake of 2.5 PgC a−1 (±0.17 PgC a−1) was found for Europe. In addition, this study reports on risks that arise from the comparison of modelled data to FLUXNET measurements and their interpretation width. Furthermore we investigate reasons for uncertainties in model results and focus here on Vmax values, and finally embed our results into a broader context of model validation studies published during the last years in order to evaluate differences or similarities in analysed error sources.

Augmentation of groundwater monitoring networks using information theory and ensemble modeling with pedotransfer functions

Summary Improving understanding of chemical transport in the subsurface commonly employs evolving groundwater monitoring networks. The objective of this work was to apply the information theory to propose an objective algorithm for augmenting a subsurface monitoring network (SMN) with the purpose of discrimination of conceptually different subsurface flow and transport models. This method determines new monitoring locations where the Kullback–Leibler total information gain is maximized. The latter is computed based on estimates of the uncertainty in modeling results and uncertainty in observations. The method was applied to discriminate models in (1) a synthetic case of groundwater contamination from a point source; (2) the tracer experiment conducted at the USDA-ARS OPE3 research site where a pulse of KCL solution was applied with irrigation water and CL − concentrations were subsequently monitored. Models were compared that included or ignored the effect of subsurface soil lenses on chemical transport. Pedotransfer functions were used to develop the ensemble of models for estimating the uncertainty in modeling results obtained with the numerical 3D flow and transport model. Peak tracer breakthrough concentrations were used to define the information gains. The determination of the new locations to augment existing ones was conducted on a 2-D grid. The information gain peaked in small area, and additional observation locations were very well spatially defined. Well-calibrated models provided a single optimal location, whereas, if models were not calibrated well, the Bayesian estimates of the new observation location depended on the activation sequence assumed for existing locations. The information gain maximization can suggest data collection locations to reduce uncertainties in the conceptual models of subsurface flow and transport.

HealthPaths—Using health trajectories to estimate and simulate the relative importance of determinants of health-adjusted life expectancy (HALE): a statistical analysis

BackgroundThere is broad agreement that a summary measure of population health such as health-adjusted life expectancy (HALE) is fundamental. While HALE is strongly correlated with income, multivariate analyses ascribing relative importance to a range of socioeconomic determinants are relatively rare. MethodsThis analysis combines very detailed statistical analysis of Statistics Canada's longitudinal National Population Health Survey (NPHS) with the HealthPaths dynamic microsimulation model. Microsimulation is used to synthesise a realistic base-case representative longitudinal population sample and then a series of carefully constructed counterfactual populations, each cutting a link in the “web of causality” estimated from the NPHS. Comparisons of the distributions of health-adjusted life lengths and summary HALE measures between counterfactuals and the base case are then used to estimate the quantitative importance of the determinants being analysed. Replicate simulations plus underlying replicate regressions further enable sophisticated measures of uncertainty in model results. FindingsThis analysis extends an earlier version that focused on four factors: obesity, smoking, education, and Antonovsky's Sense of Coherence (SoC). In that analysis, surprisingly, SoC had the largest potential impact, while education and smoking each accounted for about half as many years of (cause-deleted) HALE, and obesity accounted for relatively little. This analysis extends these results to encompass a broader range of factors, including household income, leisure and daily physical activity, employment, sense of mastery, family membership, and institutional residence. These additions change the rank order and magnitudes of the factors somewhat, reflecting omitted variable biases. The statistical analysis has had to confront a number of challenges with the main dataset, including inertia in health states, missing data, and errors in variables. InterpretationWidely observed socioeconomic gradients in health, when examined in greater detail, are the culmination of myriad co-evolving dynamic processes, where their interacting and joint effects, alone, are small. It is essential to employ an analytical framework that combines robust estimates of individual multivariate health status dynamics with health determinants dynamics in order to assess realistically the sources of health inequalities. FundingCanadian Foundation for Innovation.

Characterization of redox conditions in the excavation disturbed zone of a drift in the Kamaishi Mine, Japan

The excavation of drifts during construction of a geologic repository for high-level nuclear waste (HLW) could affect mechanical and hydraulic properties of the rock within a zone extending a short distance into the rock from drift walls. Related impacts on groundwater chemistry within such an Excavation Disturbed Zone (EDZ) are largely unknown, but the oxygen in air circulating through drifts could conceivably diffuse into groundwater within the EDZ and cause these solutions to become strongly oxidizing. A study was therefore undertaken of redox conditions within the EDZ of crystalline host rocks in the abandoned Kamaishi Fe–Cu mine in Japan, which is believed to be generally representative of conditions that could exist in the EDZ of a HLW repository. The chemical compositions of groundwaters flowing into three boreholes that were drilled various distances into the Kurihashi granodiorite were monitored continuously. The results indicated that dissolved oxygen concentrations tend to decrease with increasing distance into the rock. Oxygen penetrated, at most, a few meters into the rock from a drift (E.L. 250m drift) that intersected a specific fracture (Fracture No. 99), for example. A conceptual model was developed that takes into account both the reaction rate of dissolved oxygen with ferrous minerals and the diffusion rate of oxygen into the rock matrix. A quantitative evaluation of the model using a numerical solver indicated that the oxygen diffusion depth could vary from 3 to 30cm from the unsaturated/saturated zone boundary, depending on the reactive surface areas of ferrous minerals. These estimates may overestimate the migration distance of a redox front, however, because ferrous minerals other than biotite were not considered in the model and the partial pressure of O2(g) was fixed at the atmospheric value of 0.21bar throughout the unsaturated zone. More accurate assessments of redox-front behavior may be possible given better constraints on mineralogy and associated kinetic parameters (i.e., reactive surface areas and rate constants), and on microbial effects on oxygen consumption. Additionally, the use of more realistic boundary conditions on dissolved oxygen concentrations in the unsaturated zone would help reduce uncertainties in model results.