A score test for binary data with patient non-compliance

A score test is developed for binary clinical trial data, which incorporates patient non-compliance while respecting randomization. It is assumed in this paper that compliance is all-or-nothing, in the sense that a patient either accepts all of the treatment assigned as specified in the protocol, or none of it. Direct analytic comparisons of the adjusted test statistic for both the score test and the likelihood ratio test are made with the corresponding test statistics that adhere to the intention-to-treat principle. It is shown that no gain in power is possible over the intention-to-treat analysis, by adjusting for patient non-compliance. Sample size formulae are derived and simulation studies are used to demonstrate that the sample size approximation holds. Copyright © 2003 John Wiley & Sons, Ltd.

The severity of malarial anaemia in Plasmodium chabaudi infections of BALB/c mice is determined independently of the number of circulating parasites

Background: Severe malarial anaemia is a major complication of malaria infection and is multifactorial resulting from loss of circulating red blood cells (RBCs) from parasite replication, as well as immune-mediated mechanisms. An understanding of the causes of severe malarial anaemia is necessary to develop and implement new therapeutic strategies to tackle this syndrome of malaria infection. Methods: Using analysis of variance, this work investigated whether parasite-destruction of RBCs always accounts for the severity of malarial anaemia during infections of the rodent malaria model Plasmodium chabaudi in mice of a BALB/c background. Differences in anaemia between two different clones of P. chabaudi were also examined. Results: Circulating parasite numbers were not correlated with the severity of anaemia in either BALB/c mice or under more severe conditions of anaemia in BALB/c RAG2 deficient mice (lacking T and B cells). Mice infected with P. chabaudi clone CB suffered more severe anaemia than mice infected with clone AS, but this was not correlated with the number of parasites in the circulation. Instead, the peak percentage of parasitized RBCs was higher in CB-infected animals than in AS-infected animals, and was correlated with the severity of anaemia, suggesting that the availability of uninfected RBCs was impaired in CB-infected animals. Conclusion: This work shows that parasite numbers are a more relevant measure of parasite levels in P. chabaudi infection than % parasitaemia, a measure that does not take anaemia into account. The lack of correlation between parasite numbers and the drop in circulating RBCs in this experimental model of malaria support a role for the host response in the impairment or destruction of uninfected RBC in P. chabaudi infections, and thus development of acute anaemia in this malaria model.

Autistic symptomatology and language ability in autism spectrum disorder and specific language impairment

Background: Autism spectrum disorders (ASD) and specific language impairment (SLI) are common developmental disorders characterised by deficits in language and communication. The nature of the relationship between them continues to be a matter of debate. This study investigates whether the co-occurrence of ASD and language impairment is associated with differences in severity or pattern of autistic symptomatology or language profile. Methods: Participants (N = 97) were drawn from a total population cohort of 56,946 screened as part of study to ascertain the prevalence of ASD, aged 9 to 14 years. All children received an ICD-10 clinical diagnosis of ASD or No ASD. Children with nonverbal IQ 80 were divided into those with a language impairment (language score of 77 or less) and those without, creating three groups: children with ASD and a language impairment (ALI; N = 41), those with ASD and but no language impairment (ANL; N = 31) and those with language impairment but no ASD (SLI; N = 25). Results: Children with ALI did not show more current autistic symptoms than those with ANL. Children with SLI were well below the threshold for ASD. Their social adaptation was higher than the ASD groups, but still nearly 2 SD below average. In ALI the combination of ASD and language impairment was associated with weaker functional communication and more severe receptive language difficulties than those found in SLI. Receptive and expressive language were equally impaired in ALI, whereas in SLI receptive language was stronger than expressive. Conclusions: Co-occurrence of ASD and language impairment is not associated with increased current autistic symptomatology but appears to be associated with greater impairment in receptive language and functional communication.

Sparse kernel density construction using orthogonal forward regression with leave-one-out test score and local regularization

This paper presents an efficient construction algorithm for obtaining sparse kernel density estimates based on a regression approach that directly optimizes model generalization capability. Computational efficiency of the density construction is ensured using an orthogonal forward regression, and the algorithm incrementally minimizes the leave-one-out test score. A local regularization method is incorporated naturally into the density construction process to further enforce sparsity. An additional advantage of the proposed algorithm is that it is fully automatic and the user is not required to specify any criterion to terminate the density construction procedure. This is in contrast to an existing state-of-art kernel density estimation method using the support vector machine (SVM), where the user is required to specify some critical algorithm parameter. Several examples are included to demonstrate the ability of the proposed algorithm to effectively construct a very sparse kernel density estimate with comparable accuracy to that of the full sample optimized Parzen window density estimate. Our experimental results also demonstrate that the proposed algorithm compares favorably with the SVM method, in terms of both test accuracy and sparsity, for constructing kernel density estimates.

A combined score test for binary and ordinal endpoints from clinical trials

There is growing interest, especially for trials in stroke, in combining multiple endpoints in a single clinical evaluation of an experimental treatment. The endpoints might be repeated evaluations of the same characteristic or alternative measures of progress on different scales. Often they will be binary or ordinal, and those are the cases studied here. In this paper we take a direct approach to combining the univariate score statistics for comparing treatments with respect to each endpoint. The correlations between the score statistics are derived and used to allow a valid combined score test to be applied. A sample size formula is deduced and application in sequential designs is discussed. The method is compared with an alternative approach based on generalized estimating equations in an illustrative analysis and replicated simulations, and the advantages and disadvantages of the two approaches are discussed.

Equitability revisited: why the “equitable threat score” is not equitable

In the forecasting of binary events, verification measures that are “equitable” were defined by Gandin and Murphy to satisfy two requirements: 1) they award all random forecasting systems, including those that always issue the same forecast, the same expected score (typically zero), and 2) they are expressible as the linear weighted sum of the elements of the contingency table, where the weights are independent of the entries in the table, apart from the base rate. The authors demonstrate that the widely used “equitable threat score” (ETS), as well as numerous others, satisfies neither of these requirements and only satisfies the first requirement in the limit of an infinite sample size. Such measures are referred to as “asymptotically equitable.” In the case of ETS, the expected score of a random forecasting system is always positive and only falls below 0.01 when the number of samples is greater than around 30. Two other asymptotically equitable measures are the odds ratio skill score and the symmetric extreme dependency score, which are more strongly inequitable than ETS, particularly for rare events; for example, when the base rate is 2% and the sample size is 1000, random but unbiased forecasting systems yield an expected score of around −0.5, reducing in magnitude to −0.01 or smaller only for sample sizes exceeding 25 000. This presents a problem since these nonlinear measures have other desirable properties, in particular being reliable indicators of skill for rare events (provided that the sample size is large enough). A potential way to reconcile these properties with equitability is to recognize that Gandin and Murphy’s two requirements are independent, and the second can be safely discarded without losing the key advantages of equitability that are embodied in the first. This enables inequitable and asymptotically equitable measures to be scaled to make them equitable, while retaining their nonlinearity and other properties such as being reliable indicators of skill for rare events. It also opens up the possibility of designing new equitable verification measures.

Impacts of changing air composition on severity of arable crop disease epidemics

This review assesses the impacts, both direct and indirect, of man-made changes to the composition of the air over a 200 year period on the severity of arable crop disease epidemics. The review focuses on two well-studied UK arable crops,wheat and oilseed rape, relating these examples to worldwide food security. In wheat, impacts of changes in concentrations of SO2 in air on two septoria diseases are discussed using data obtained from historical crop samples and unpublished experimental work. Changes in SO2 seem to alter septoria disease spectra both through direct effects on infection processes and through indirect effects on soil S status. Work on the oilseed rape diseases phoma stem canker and light leaf spot illustrates indirect impacts of increasing concentrations of greenhouse gases, mediated through climate change. It is projected that, by the 2050s, if diseases are not controlled, climate change will increase yields in Scotland but halve yields in southern England. These projections are discussed in relation to strategies for adaptation to environmental change. Since many strategies take10–15 years to implement, it is important to take appropriate decisions soon. Furthermore, it is essential to make appropriate investment in collation of long-term data, modelling and experimental work to guide such decision-making by industry and government, as a contribution to worldwide food security.

A method for the economic valuation of animal welfare benefits using a single welfare score

Unless the benefits to society of measures to protect and improve the welfare of animals are made transparent by means of their valuation they are likely to go unrecognised and cannot easily be weighed against the costs of such measures as required, for example, by policy-makers. A simple single measure scoring system, based on the Welfare Quality® index, is used, together with a choice experiment economic valuation method, to estimate the value that people place on improvements to the welfare of different farm animal species measured on a continuous (0-100) scale. Results from using the method on a survey sample of some 300 people show that it is able to elicit apparently credible values. The survey found that 96% of respondents thought that we have a moral obligation to safeguard the welfare of animals and that over 72% were concerned about the way farm animals are treated. Estimated mean annual willingness to pay for meat from animals with improved welfare of just one point on the scale was £5.24 for beef cattle, £4.57 for pigs and £5.10 for meat chickens. Further development of the method is required to capture the total economic value of animal welfare benefits. Despite this, the method is considered a practical means for obtaining economic values that can be used in the cost-benefit appraisal of policy measures intended to improve the welfare of animals.

An item response analysis of the motor and behavioral subscales of the unified Huntington's disease rating scale in huntington disease gene expansion carriers

Although the Unified Huntington's Disease Rating Scale (UHDRS) is widely used in the assessment of Huntington disease (HD), the ability of individual items to discriminate individual differences in motor or behavioral manifestations has not been extensively studied in HD gene expansion carriers without a motor-defined clinical diagnosis (ie, prodromal-HD or prHD). To elucidate the relationship between scores on individual motor and behavioral UHDRS items and total score for each subscale, a nonparametric item response analysis was performed on retrospective data from 2 multicenter longitudinal studies. Motor and behavioral assessments were supplied for 737 prHD individuals with data from 2114 visits (PREDICT-HD) and 686 HD individuals with data from 1482 visits (REGISTRY). Option characteristic curves were generated for UHDRS subscale items in relation to their subscale score. In prHD, overall severity of motor signs was low, and participants had scores of 2 or above on very few items. In HD, motor items that assessed ocular pursuit, saccade initiation, finger tapping, tandem walking, and to a lesser extent, saccade velocity, dysarthria, tongue protrusion, pronation/supination, Luria, bradykinesia, choreas, gait, and balance on the retropulsion test were found to discriminate individual differences across a broad range of motor severity. In prHD, depressed mood, anxiety, and irritable behavior demonstrated good discriminative properties. In HD, depressed mood demonstrated a good relationship with the overall behavioral score. These data suggest that at least some UHDRS items appear to have utility across a broad range of severity, although many items demonstrate problematic features.

Estimating reliability and resolution of probability forecasts through decomposition of the empirical score

References (20)Cited By (1)Export CitationAboutAbstract Proper scoring rules provide a useful means to evaluate probabilistic forecasts. Independent from scoring rules, it has been argued that reliability and resolution are desirable forecast attributes. The mathematical expectation value of the score allows for a decomposition into reliability and resolution related terms, demonstrating a relationship between scoring rules and reliability/resolution. A similar decomposition holds for the empirical (i.e. sample average) score over an archive of forecast–observation pairs. This empirical decomposition though provides a too optimistic estimate of the potential score (i.e. the optimum score which could be obtained through recalibration), showing that a forecast assessment based solely on the empirical resolution and reliability terms will be misleading. The differences between the theoretical and empirical decomposition are investigated, and specific recommendations are given how to obtain better estimators of reliability and resolution in the case of the Brier and Ignorance scoring rule.

Evaluating raw ensembles with the continuous ranked probability score

The continuous ranked probability score (CRPS) is a frequently used scoring rule. In contrast with many other scoring rules, the CRPS evaluates cumulative distribution functions. An ensemble of forecasts can easily be converted into a piecewise constant cumulative distribution function with steps at the ensemble members. This renders the CRPS a convenient scoring rule for the evaluation of ‘raw’ ensembles, obviating the need for sophisticated ensemble model output statistics or dressing methods prior to evaluation. In this article, a relation between the CRPS score and the quantile score is established. The evaluation of ‘raw’ ensembles using the CRPS is discussed in this light. It is shown that latent in this evaluation is an interpretation of the ensemble as quantiles but with non-uniform levels. This needs to be taken into account if the ensemble is evaluated further, for example with rank histograms.

Developing consensus on hospital prescribing indicators of potential harms amenable to decision support

Aim: To develop a list of prescribing indicators specific for the hospital setting that would facilitate the prospective collection of high severity and/or high frequency prescribing errors, which are also amenable to electronic clinical decision support (CDS). Method: A three-stage consensus technique (electronic Delphi) was carried out with 20 expert pharmacists and physicians across England. Participants were asked to score prescribing errors using a 5-point Likert scale for their likelihood of occurrence and the severity of the most likely outcome. These were combined to produce risk scores, from which median scores were calculated for each indicator across the participants in the study. The degree of consensus between the participants was defined as the proportion that gave a risk score in the same category as the median. Indicators were included if a consensus of 80% or more was achieved. Results: A total of 80 prescribing errors were identified by consensus as being high or extreme risk. The most common drug classes named within the indicators were antibiotics (n=13), antidepressants (n=8), nonsteroidal anti-inflammatory drugs (n=6), and opioid analgesics (n=6).The most frequent error type identified as high or extreme risk were those classified as clinical contraindications (n=29/80). Conclusion: 80 high risk prescribing errors in the hospital setting have been identified by an expert panel. These indicators can serve as the basis for a standardised, validated tool for the collection of data in both paperbased and electronic prescribing processes, as well as to assess the impact of electronic decision support implementation or development.

Accounting for the economic risk, caused by variation in disease severity, in fungicide dose decisions: exemplified for Mycosphaerella graminicola on winter wheat

A method is presented to calculate economic optimum fungicide doses accounting for the risk-aversion of growers responding to variability in disease severity between crops. Simple dose-response and disease-yield loss functions are used to estimate net disease-related costs (fungicide cost, plus disease-induced yield loss) as a function of dose and untreated severity. With fairly general assumptions about the shapes of the probability distribution of disease severity and the other functions involved, we show that a choice of fungicide dose which minimises net costs on average across seasons results in occasional large net costs caused by inadequate control in high disease seasons. This may be unacceptable to a grower with limited capital. A risk-averse grower can choose to reduce the size and frequency of such losses by applying a higher dose as insurance. For example, a grower may decide to accept ‘high loss’ years one year in ten or one year in twenty (i.e. specifying a proportion of years in which disease severity and net costs will be above a specified level). Our analysis shows that taking into account disease severity variation and risk-aversion will usually increase the dose applied by an economically rational grower. The analysis is illustrated with data on septoria tritici leaf blotch of wheat caused by Mycosphaerella graminicola. Observations from untreated field plots at sites across England over three years were used to estimate the probability distribution of disease severities at mid-grain filling. In the absence of a fully reliable disease forecasting scheme, reducing the frequency of ‘high loss’ years requires substantially higher doses to be applied to all crops. Disease resistant cultivars reduce both the optimal dose at all levels of risk and the disease-related costs at all doses.