Active-control trials with binary data: a comparison of methods for testing superiority or non-inferiority using the odds ratio

This paper considers methods for testing for superiority or non-inferiority in active-control trials with binary data, when the relative treatment effect is expressed as an odds ratio. Three asymptotic tests for the log-odds ratio based on the unconditional binary likelihood are presented, namely the likelihood ratio, Wald and score tests. All three tests can be implemented straightforwardly in standard statistical software packages, as can the corresponding confidence intervals. Simulations indicate that the three alternatives are similar in terms of the Type I error, with values close to the nominal level. However, when the non-inferiority margin becomes large, the score test slightly exceeds the nominal level. In general, the highest power is obtained from the score test, although all three tests are similar and the observed differences in power are not of practical importance. Copyright (C) 2007 John Wiley & Sons, Ltd.

A nonparametric ensemble transform method for Bayesian inference

Many applications, such as intermittent data assimilation, lead to a recursive application of Bayesian inference within a Monte Carlo context. Popular data assimilation algorithms include sequential Monte Carlo methods and ensemble Kalman filters (EnKFs). These methods differ in the way Bayesian inference is implemented. Sequential Monte Carlo methods rely on importance sampling combined with a resampling step, while EnKFs utilize a linear transformation of Monte Carlo samples based on the classic Kalman filter. While EnKFs have proven to be quite robust even for small ensemble sizes, they are not consistent since their derivation relies on a linear regression ansatz. In this paper, we propose another transform method, which does not rely on any a priori assumptions on the underlying prior and posterior distributions. The new method is based on solving an optimal transportation problem for discrete random variables. © 2013, Society for Industrial and Applied Mathematics

Online bayesian inference in some time-frequency representations of non-stationary processes

The use of Bayesian inference in the inference of time-frequency representations has, thus far, been limited to offline analysis of signals, using a smoothing spline based model of the time-frequency plane. In this paper we introduce a new framework that allows the routine use of Bayesian inference for online estimation of the time-varying spectral density of a locally stationary Gaussian process. The core of our approach is the use of a likelihood inspired by a local Whittle approximation. This choice, along with the use of a recursive algorithm for non-parametric estimation of the local spectral density, permits the use of a particle filter for estimating the time-varying spectral density online. We provide demonstrations of the algorithm through tracking chirps and the analysis of musical data.

Skill Rating by Bayesian Inference

Systems Engineering often involves computer modelling the behaviour of proposed systems and their components. Where a component is human, fallibility must be modelled by a stochastic agent. The identification of a model of decision-making over quantifiable options is investigated using the game-domain of Chess. Bayesian methods are used to infer the distribution of players’ skill levels from the moves they play rather than from their competitive results. The approach is used on large sets of games by players across a broad FIDE Elo range, and is in principle applicable to any scenario where high-value decisions are being made under pressure.

How large is congressional dependence in agriculture. Bayesian inference about ‘scale’ and ‘scope’ in measuring a spatial externality

The political economy literature on agriculture emphasizes influence over political outcomes via lobbying conduits in general, political action committee contributions in particular and the pervasive view that political preferences with respect to agricultural issues are inherently geographic. In this context, ‘interdependence’ in Congressional vote behaviour manifests itself in two dimensions. One dimension is the intensity by which neighboring vote propensities influence one another and the second is the geographic extent of voter influence. We estimate these facets of dependence using data on a Congressional vote on the 2001 Farm Bill using routine Markov chain Monte Carlo procedures and Bayesian model averaging, in particular. In so doing, we develop a novel procedure to examine both the reliability and the consequences of different model representations for measuring both the ‘scale’ and the ‘scope’ of spatial (geographic) co-relations in voting behaviour.

Sample size considerations in active-control non-inferiority trials with binary data based on the odds ratio

This paper presents an approximate closed form sample size formula for determining non-inferiority in active-control trials with binary data. We use the odds-ratio as the measure of the relative treatment effect, derive the sample size formula based on the score test and compare it with a second, well-known formula based on the Wald test. Both closed form formulae are compared with simulations based on the likelihood ratio test. Within the range of parameter values investigated, the score test closed form formula is reasonably accurate when non-inferiority margins are based on odds-ratios of about 0.5 or above and when the magnitude of the odds ratio under the alternative hypothesis lies between about 1 and 2.5. The accuracy generally decreases as the odds ratio under the alternative hypothesis moves upwards from 1. As the non-inferiority margin odds ratio decreases from 0.5, the score test closed form formula increasingly overestimates the sample size irrespective of the magnitude of the odds ratio under the alternative hypothesis. The Wald test closed form formula is also reasonably accurate in the cases where the score test closed form formula works well. Outside these scenarios, the Wald test closed form formula can either underestimate or overestimate the sample size, depending on the magnitude of the non-inferiority margin odds ratio and the odds ratio under the alternative hypothesis. Although neither approximation is accurate for all cases, both approaches lead to satisfactory sample size calculation for non-inferiority trials with binary data where the odds ratio is the parameter of interest.

Statistical inference of OH concentrations and air mass dilution rates from successive observations of non-methane hydrocarbons in single air masses

Bayesian inference has been used to determine rigorous estimates of hydroxyl radical concentrations () and air mass dilution rates (K) averaged following air masses between linked observations of nonmethane hydrocarbons (NMHCs) spanning the North Atlantic during the Intercontinental Transport and Chemical Transformation (ITCT)-Lagrangian-2K4 experiment. The Bayesian technique obtains a refined (posterior) distribution of a parameter given data related to the parameter through a model and prior beliefs about the parameter distribution. Here, the model describes hydrocarbon loss through OH reaction and mixing with a background concentration at rate K. The Lagrangian experiment provides direct observations of hydrocarbons at two time points, removing assumptions regarding composition or sources upstream of a single observation. The estimates are sharpened by using many hydrocarbons with different reactivities and accounting for their variability and measurement uncertainty. A novel technique is used to construct prior background distributions of many species, described by variation of a single parameter . This exploits the high correlation of species, related by the first principal component of many NMHC samples. The Bayesian method obtains posterior estimates of , K and following each air mass. Median values are typically between 0.5 and 2.0 × 106 molecules cm−3, but are elevated to between 2.5 and 3.5 × 106 molecules cm−3, in low-level pollution. A comparison of estimates from absolute NMHC concentrations and NMHC ratios assuming zero background (the “photochemical clock” method) shows similar distributions but reveals systematic high bias in the estimates from ratios. Estimates of K are ∼0.1 day−1 but show more sensitivity to the prior distribution assumed.

Performance and prediction: Bayesian modelling of fallible choice in chess

Evaluating agents in decision-making applications requires assessing their skill and predicting their behaviour. Both are well developed in Poker-like situations, but less so in more complex game and model domains. This paper addresses both tasks by using Bayesian inference in a benchmark space of reference agents. The concepts are explained and demonstrated using the game of chess but the model applies generically to any domain with quantifiable options and fallible choice. Demonstration applications address questions frequently asked by the chess community regarding the stability of the rating scale, the comparison of players of different eras and/or leagues, and controversial incidents possibly involving fraud. The last include alleged under-performance, fabrication of tournament results, and clandestine use of computer advice during competition. Beyond the model world of games, the aim is to improve fallible human performance in complex, high-value tasks.

Productivity and optimum plant density of pigeonpea in different environments in Tanzania

The objective of the present study was to determine the optimum plant density of four pigeonpea genotypes, representing early, medium and late maturing types, grown in five contrasting environments in Tanzania. ICPL 86005 (early), Kat 50/3 and QP 37 (medium) and Local (late) were grown at four plant densities (40 000-320 000 plants/ha) in irrigated and rainfed conditions at Ilonga and under rainfed conditions at Kibaha, Selian and Ismani. At maturity, total above-ground biomass and seed yield (SY) were measured. The highest yields were obtained in the irrigated experiment at Ilonga, where the medium/late genotypes produced 25 t biomass/ha and 5 center dot 6 t seed/ha. The lowest SY were at Kibaha, 0 58 to 1 center dot 76 t/ha, where a severe drought occurred. In nearly all cases the response to density was linear or asymptotic. The response of ICPL 86005 was significantly different from the other three genotypes. The optimum density for SY varied from 37 000 to 227 000 plants/ha in ICPL 86005, compared with 3000 to 101000 plants/ha in the medium/late genotypes. The highest optimum density was at Selian and Ismani and the lowest at Ilonga and Kibaha, where drought occurred. Optimum densities therefore varied greatly with genotype (duration) and environment, and this variation needs to be considered when planning trials.

The acceptability of isoflavones as a treatment of menopausal symptoms: a European survey among postmenopausal women

Objective: To investigate determinants of the acceptability of isoflavone products among postmenopausal women with regard to social and lifestyle factors, dietary habits, health concerns, food beliefs, menopausal symptoms and therapies, and to elucidate preferences for specific products. Methods: A consumer survey was conducted among postmenopausal women in four European countries (Germany, Denmark, Italy and the UK), including a total of 465 respondents. Results: The declared acceptability of isoflavones was highest in Germany (80%), followed by Italy (75%), the UK (59%) and Denmark (55%; p < 0.001). Among other factors, prior information about isoflavones (odds ratio (OR) 2.64, 95% confidence interval (CI) 1.54-4.51, p < 0.001) and belief that foods can be used to treat medical conditions (OR 2.18, 95% CI 1.17-4.05, p=0.014) were the strongest predictors of potential acceptability. The preferred forms in which isoflavones might be taken differed markedly between countries. Conclusions: Information about the health effects of isoflavones and health-diet links should be considered the most important factors in attempts to influence the potential acceptability of isoflavone products. Foods enriched with isoflavones need to be developed with regard to specific cultural demands in order to assure consumer acceptance on the marketplace.

The effects of pre- and postnatal depression in fathers: a natural experiment comparing the effects of exposure to depression on offspring

Background: Depression in fathers in the postnatal period is associated with an increased risk of behavioural problems in their offspring, particularly for boys. The aim of this study was to examine for differential effects of depression in fathers on children's subsequent psychological functioning via a natural experiment comparing prenatal and postnatal exposure. Methods:In a longitudinal population cohort study (the Avon Longitudinal Study of Parents and Children (ALSPAC)) we examined the associations between depression in fathers measured in the prenatal and postnatal period (measured using the Edinburgh Postnatal Depression Scale), and later behavioural/emotional and psychiatric problems in their children, assessed at ages 31/2 and 7 years. Results: Children whose fathers were depressed in both the prenatal and postnatal periods had the highest risks of subsequent psychopathology, measured by total problems at age 31/2 years (Odds Ratio 3.55; 95% confidence interval 2.07, 6.08) and psychiatric diagnosis at age 7 years (OR 2.54; 1.19, 5.41). Few differences emerged when prenatal and postnatal depression exposure were directly compared, but when compared to fathers who were not depressed, boys whose fathers had postnatal depression only had higher rates of conduct problems aged 31/2 years (OR 2.14; 1.22, 3.72) whereas sons of the prenatal group did not (OR 1.41; .75, 2.65). These associations changed little when controlling for maternal depression and other potential confounding factors. Conclusions: The findings of this study suggest that the increased risk of later conduct problems, seen particularly in the sons of depressed fathers, maybe partly mediated through environmental means. In addition, children whose fathers are more chronically depressed appear to be at a higher risk of emotional and behavioural problems. Efforts to identify the precise mechanisms by which transmission of risk may occur should be encouraged to enable the development of focused interventions to mitigate risks for young children.

Possible role of aerosol transmission in a hospital outbreak of influenza.

BACKGROUND: We examined the role of aerosol transmission of influenza in an acute ward setting. METHODS: We investigated a seasonal influenza A outbreak that occurred in our general medical ward (with open bay ward layout) in 2008. Clinical and epidemiological information was collected in real time during the outbreak. Spatiotemporal analysis was performed to estimate the infection risk among patients. Airflow measurements were conducted, and concentrations of hypothetical virus-laden aerosols at different ward locations were estimated using computational fluid dynamics modeling. RESULTS: Nine inpatients were infected with an identical strain of influenza A/H3N2 virus. With reference to the index patient's location, the attack rate was 20.0% and 22.2% in the "same" and "adjacent" bays, respectively, but 0% in the "distant" bay (P = .04). Temporally, the risk of being infected was highest on the day when noninvasive ventilation was used in the index patient; multivariate logistic regression revealed an odds ratio of 14.9 (95% confidence interval, 1.7-131.3; P = .015). A simultaneous, directional indoor airflow blown from the "same" bay toward the "adjacent" bay was found; it was inadvertently created by an unopposed air jet from a separate air purifier placed next to the index patient's bed. Computational fluid dynamics modeling revealed that the dispersal pattern of aerosols originated from the index patient coincided with the bed locations of affected patients. CONCLUSIONS: Our findings suggest a possible role of aerosol transmission of influenza in an acute ward setting. Source and engineering controls, such as avoiding aerosol generation and improving ventilation design, may warrant consideration to prevent nosocomial outbreaks.

Skilloscopy: Bayesian modeling of decision makers' skill

This paper proposes and demonstrates an approach, Skilloscopy, to the assessment of decision makers. In an increasingly sophisticated, connected and information-rich world, decision making is becoming both more important and more difficult. At the same time, modelling decision-making on computers is becoming more feasible and of interest, partly because the information-input to those decisions is increasingly on record. The aims of Skilloscopy are to rate and rank decision makers in a domain relative to each other: the aims do not include an analysis of why a decision is wrong or suboptimal, nor the modelling of the underlying cognitive process of making the decisions. In the proposed method a decision-maker is characterised by a probability distribution of their competence in choosing among quantifiable alternatives. This probability distribution is derived by classic Bayesian inference from a combination of prior belief and the evidence of the decisions. Thus, decision-makers’ skills may be better compared, rated and ranked. The proposed method is applied and evaluated in the gamedomain of Chess. A large set of games by players across a broad range of the World Chess Federation (FIDE) Elo ratings has been used to infer the distribution of players’ rating directly from the moves they play rather than from game outcomes. Demonstration applications address questions frequently asked by the Chess community regarding the stability of the Elo rating scale, the comparison of players of different eras and/or leagues, and controversial incidents possibly involving fraud. The method of Skilloscopy may be applied in any decision domain where the value of the decision-options can be quantified.

Association of lipoprotein lipase gene polymorphisms with obesity and type 2 diabetes in an Asian Indian population

AIMS: Lipoprotein lipase (LPL), a pivotal enzyme in lipoprotein metabolism, catalyzes the hydrolysis of triglycerides of very low-density lipoproteins and chylomicrons. Assuming that the variants in the promoter of the LPL gene may be associated with changes in lipid metabolism leading to obesity and type 2 diabetes, we examined the role of promoter variants (-T93G and -G53C) in the LPL gene in an urban South Indian population. METHODS: The study subjects (619 type 2 diabetic and 731 normal glucose-tolerant (NGT) subjects) were chosen from the Chennai Urban Rural Epidemiology Study, an ongoing population-based study in southern India. The polymorphisms were genotyped using polymerase chain reaction-restriction-fragment length polymorphism (PCR-RFLP). Linkage disequilibrium (LD) was estimated from the estimates of haplotypic frequencies. RESULTS: The two polymorphisms studied were not in LD. The -T93G was not associated with type 2 diabetes but was associated with obesity. 11.5% of the obese subjects (62/541) had the XG(TG+GG) genotype compared with 6.4% of the nonobese subjects (52/809; P=0.001). The odds ratio for obesity for the XG genotype was 1.766 (95% CI: 1.19-2.63, P=0.005). Subjects with XG genotype also had higher body mass index and waist circumference compared with those with TT genotype. With respect to G53C, subjects with the XC(GC+CC) genotype had 0.527 and 0.531 times lower risk for developing type 2 diabetes and obesity, respectively. CONCLUSIONS: Among Asian Indians, the -T93G SNP of the LPL gene is associated with obesity but not type 2 diabetes, whereas the -G53C SNP appears to be protective against both obesity and type 2 diabetes.

Thr54 allele carriers of the Ala54Thr variant of FABP2 gene have associations with metabolic syndrome and hypertriglyceridemia in urban South Indians

The intestinal fatty acid-binding protein gene is proposed as a candidate gene for diabetes because the protein it codes is involved in fatty acid absorption and metabolism. This study investigates the association of the Ala54Thr variant of the intestinal fatty acid-binding protein gene on type 2 diabetes mellitus and other related metabolic traits in Asian Indians. Ala54Thr polymorphism was genotyped by using polymerase chain reaction-restriction fragment length polymorphism in unrelated 773 type 2 diabetic and 899 normal glucose-tolerant (NGT) subjects, randomly chosen from the Chennai Urban Rural Epidemiology Study, an ongoing population-based study in South India. The Ala54Thr polymorphism was not associated with type 2 diabetes mellitus or obesity. However, genotype-phenotype study revealed that the NGT subjects carrying the Thr54 allele had significantly higher 2-hour plasma glucose (P = .007), glycated hemoglobin (P = .004), 2-hour insulin (P = .027), and fasting low-density lipoprotein cholesterol (P = .032) levels compared with those with the Ala54 allele. Normal glucose-tolerant subjects with Ala54Thr and Thr54Thr genotypes had significantly higher fasting serum triglyceride levels (P = .003) compared with those with Ala54Ala. The subjects were stratified into those with hypertriglyceridemia (serum triglyceride levels >or=150 mg/dL) and those without. The odds ratio for hypertriglyceridemia for the individuals carrying the Ala54Thr genotype was 1.491 (95% confidence interval [CI], 1.22-1.83, P < .0001), and for those carrying the Thr54Thr genotype, it was 1.888 (95% CI, 1.34-2.67; P < .0001). Subjects were also stratified into those with metabolic syndrome (MS) and those without, according to modified Adult Treatment Panel III guidelines. The odds ratio (adjusted for age and sex) for MS for the individuals carrying the Ala54Thr genotype was 1.240 (95% CI, 1.02-1.51; P = .03), whereas for those carrying the Thr54Thr genotype, it was 1.812 (95% CI, 1.28-2.57; P = .001). Carriers of the Thr54 allele have associations with MS and hypertriglyceridemia in this urban South Indian population.