A Note on Combining Dependent Tests of Significance

In combining several tests of significance the individual test statistics are allowed to be stochastically dependent. By choosing the weighted inverse normal method for the combination, the dependency of the original test statistics is then characterized by a correlation of the transformed statistics. For this correlation a confidence region, an unbiased estimator and an unbiased estimate of its variance are derived. The combined test statistic is extended to include the case of possibly dependent original test statistics. Simulation studies show the performance of the actual significance level.     

Bayesian Meta‐Analysis of the Accuracy of a Test for Tuberculous Pleuritis in the Absence of a Gold Standard Reference

Summary Absence of a perfect reference test is an acknowledged source of bias in diagnostic studies. In the case of tuberculous pleuritis, standard reference tests such as smear microscopy, culture and biopsy have poor sensitivity. Yet meta‐analyses of new tests for this disease have always assumed the reference standard is perfect, leading to biased estimates of the new test’s accuracy. We describe a method for joint meta‐analysis of sensitivity and specificity of the diagnostic test under evaluation, while considering the imperfect nature of the reference standard. We use a Bayesian hierarchical model that takes into account within‐ and between‐study variability. We show how to obtain pooled estimates of sensitivity and specificity, and how to plot a hierarchical summary receiver operating characteristic curve. We describe extensions of the model to situations where multiple reference tests are used, and where index and reference tests are conditionally dependent. The performance of the model is evaluated using simulations and illustrated using data from a meta‐analysis of nucleic acid amplification tests (NAATs) for tuberculous pleuritis. The estimate of NAAT specificity was higher and the sensitivity lower compared to a model that assumed that the reference test was perfect.     

Sensitivity and Validity of Psychometric Tests for Assessing Driving Impairment: Effects of Sleep Deprivation

ObjectiveTo assess drug induced driving impairment, initial screening is needed. However, no consensus has been reached about which initial screening tools have to be used. The present study aims to determine the ability of a battery of psychometric tests to detect performance impairing effects of clinically relevant levels of drowsiness as induced by one night of sleep deprivation.MethodsTwenty four healthy volunteers participated in a 2-period crossover study in which the highway driving test was conducted twice: once after normal sleep and once after one night of sleep deprivation. The psychometric tests were conducted on 4 occasions: once after normal sleep (at 11 am) and three times during a single night of sleep deprivation (at 1 am, 5 am, and 11 am).ResultsOn-the-road driving performance was significantly impaired after sleep deprivation, as measured by an increase in Standard Deviation of Lateral Position (SDLP) of 3.1 cm compared to performance after a normal night of sleep. At 5 am, performance in most psychometric tests showed significant impairment. As expected, largest effect sizes were found on performance in the Psychomotor Vigilance Test (PVT). Large effects sizes were also found in the Divided Attention Test (DAT), the Attention Network Test (ANT), and the test for Useful Field of View (UFOV) at 5 and 11 am during sleep deprivation. Effects of sleep deprivation on SDLP correlated significantly with performance changes in the PVT and the DAT, but not with performance changes in the UFOV.ConclusionFrom the psychometric tests used in this study, the PVT and DAT seem most promising for initial evaluation of drug impairment based on sensitivity and correlations with driving impairment. Further studies are needed to assess the sensitivity and validity of these psychometric tests after benchmark sedative drug use.     

Human cognitive performance in a 3 mT power-line frequency magnetic field

Extremely low frequency (ELF, <300 Hz) magnetic fields (MF) have been reported to modulate cognitive performance in humans. However, little research exists with MF exposures comparable to the highest levels experienced in occupations like power line workers and industrial welders. This research aims to evaluate the impact of a 60 Hz, 3 mT MF on human cognitive performance. Ninety‐nine participants completed the double‐blind protocol, performing a selection of psychometric tests under two consecutive MF exposure conditions dictated by assignment to one of three groups (sham/sham, MF exposure/sham, or sham/MF exposure). Data were analyzed using a 3 × 2 mixed model analysis of variance. Performance between repetitions improved in 11 of 15 psychometric parameters (practice effect). A significant interaction effect on the digit span forward test (F = 5.21, P < 0.05) revealed an absence of practice effects for both exposure groups but not the control group. This memory test indicates MF‐induced abolition of the improvement associated with practice. Overall, this study does not establish any clear MF effect on human cognition. It is speculated that an ELF MF may interfere with the neuropsychological processes responsible for this short‐term learning effect supported by brain synaptic plasticity. Bioelectromagnetics. Bioelectromagnetics 32:620–633, 2011. © 2011 Wiley Periodicals, Inc.     

Relationship between Measures of Exposure to PCBs/Dioxins and Behavioral Effects in Recent Developmental Studies

This paper does not reflect official EPA policy. Epidemiological studies on the neurodevelopmental effects of exposure to PCBs initiated in the last decade have had the opportunity to take advantage of modern methodologies for the analysis of congeners of PCBs, dioxins, and related orga-nochlorine compounds. Each of these studies is a longitudinal prospective study, in which women were recruited during pregnancy and the children are being followed for at least several years after birth. The study from which the largest body of data has been published to date is being performed in the Netherlands, in which exposure to PCBs and related compounds is through the general food supply. Mother-infant pairs were recruited in two cities. Half of the infants were bottle-fed and half breast-fed in each city. Four PCB congeners (118, 138, 153, 180) were assessed in maternal and cord plasma, breast milk, and plasma of the child at 3.5 years. TEQ in breast milk was calculated based on PCDDs/PCDFs and dioxin-like PCBs. Various measures of in utero exposure were associated with suboptimal neurological status during infancy, whereas maternal plasma PCB concentration was associated with cognitive deficits (Kaufman scores) at 3.5 years of age. The child's concurrent plasma PCB levels and maternal PCB plasma levels independently predicted performance on various aspects of a vigilance task, and maternal and cord plasma levels predicted impairment of complex play behavior. Poor scores on behavioral ratings were associated with concurrent blood PCB concentrations in the child. A study in Oswego in Lake Ontario fish eaters includes mothers who never ate Great Lakes fish and mothers who consumed greater than 40 PCB-equivalent pounds of Lake Ontario fish over their lifetime. Sixty-eight PCB congeners were measured in cord blood. Suboptimal neurological status during infancy was associated with maternal fish consumption and highly chlorinated cord PCB levels, whereas deficits in short-term memory at 6 months and 1 year of age were associated with total chlorinated cord PCB levels. In a study in Germany of 171 mother-infant pairs, PCB congeners 138, 153, and 180 were measured in cord plasma and milk 2 weeks after birth; both measures are considered markers of in utero exposure. Suboptimal neurological status during infancy, decrements in Bayley scores at 30 months and Kaufman scores at 42 months were associated with PCBs in milk but not cord plasma. These studies, combined with data from previous studies, reveal a consistent relationship between PCB exposure and suboptimal neurological status during infancy, and cognitive deficits associated with in utero exposure. Data from the Dutch study revealed effects on other behavioral domains associated with concurrent (postnatal) exposure. Although it is not possible to identify specific congeners or groups of congeners that may be responsible for the neurotoxic effects observed in these studies, the TEQ approach was not particularly predictive for neurotoxic outcomes.     

Pertussis immunisation and serious acute neurological illnesses in children.

OBJECTIVE--To determine long term outcome in children who had a severe acute neurological illness in early childhood associated with pertussis immunisation. DESIGN--Follow up study of cases and matched controls. SETTING--Assessment of children at home and at school throughout Britain. SUBJECTS--Children recruited into the national childhood encephalopathy study in 1976-9 were followed up, with one of their two original matched controls, in 1986-9. MAIN OUTCOME MEASURES--Performance in educational attainment tests; behaviour problems reported by teachers and parents; continuing convulsions; evidence of other neurological or physical dysfunction. RESULTS--Over 80% of cases and controls were traced. Case children were significantly more likely than controls to have died or to have some form of educational, behavioural, neurological, or physical dysfunction a decade after their illness. The prevalence of one or more of these adverse outcomes in case children who had been immunised with diphtheria, tetanus, and pertussis vaccine within seven days before onset of their original illness was similar to that in case children who had not been immunised recently. The relative risk for recent diphtheria, tetanus, and pertussis immunisation in children who had died or had any dysfunction in comparison with controls was 5.5 (95% confidence interval 1.6 to 23.7). However, the number of cases associated with vaccine (12) was extremely small and statistically vulnerable, and other possible agents or predisposing factors could not be excluded. CONCLUSIONS--Diphtheria, tetanus, and pertussis vaccine may on rare occasions be associated with the development of severe acute neurological illnesses that can have serious sequelae. Some cases may occur by chance or have other causes. The role of pertussis vaccine as a prime or concomitant factor in the aetiology of these illnesses cannot be determined in any individual case. The balance of possible risk against known benefits from pertussis immunisation supports continued use of the vaccine.     

Scaled marginal models for multiple continuous outcomes

In studies that involve multivariate outcomes it is often of interest to test for a common exposure effect. For example, our research is motivated by a study of neurocognitive performance in a cohort of HIV-infected women. The goal is to determine whether highly active antiretroviral therapy affects different aspects of neurocognitive functioning to the same degree and if so, to test for the treatment effect using a more powerful one-degree-of-freedom global test. Since multivariate continuous outcomes are likely to be measured on different scales, such a common exposure effect has not been well defined. We propose the use of a scaled marginal model for testing and estimating this global effect when the outcomes are all continuous. A key feature of the model is that the effect of exposure is represented by a common effect size and hence has a well-understood, practical interpretation. Estimating equations are proposed to estimate the regression coefficients and the outcome-specific scale parameters, where the correct specification of the within-subject correlation is not required. These estimating equations can be solved by repeatedly calling standard generalized estimating equations software such as SAS PROC GENMOD. To test whether the assumption of a common exposure effect is reasonable, we propose the use of an estimating-equation-based score-type test. We study the asymptotic efficiency loss of the proposed estimators, and show that they generally have high efficiency compared to the maximum likelihood estimators. The proposed method is applied to the HIV data.     

Predicting Performance in Higher Education Using Proximal Predictors

We studied the validity of two methods for predicting academic performance and student-program fit that were proximal to important study criteria. Applicants to an undergraduate psychology program participated in a selection procedure containing a trial-studying test based on a work sample approach, and specific skills tests in English and math. Test scores were used to predict academic achievement and progress after the first year, achievement in specific course types, enrollment, and dropout after the first year. All tests showed positive significant correlations with the criteria. The trial-studying test was consistently the best predictor in the admission procedure. We found no significant differences between the predictive validity of the trial-studying test and prior educational performance, and substantial shared explained variance between the two predictors. Only applicants with lower trial-studying scores were significantly less likely to enroll in the program. In conclusion, the trial-studying test yielded predictive validities similar to that of prior educational performance and possibly enabled self-selection. In admissions aimed at student-program fit, or in admissions in which past educational performance is difficult to use, a trial-studying test is a good instrument to predict academic performance.     

In vivo validation of in vitro quality tests for cryopreserved honey bee semen

Development of cryopreservation protocols for honey bee semen is hampered by the lack of validated laboratory tests that allow the prediction of in vivo performance of frozen-thawed semen. Here we analyzed correlations between seven in vitro tests and indicators of semen performance after insemination. These tests included measures of motility, cell conformation, and membrane permeability before and after exposure to physiochemical stress. We show that the proposed protocol for motility measurement yields results that correlate well with the number of sperm reaching the storage organ of queens (correlation coefficient ρ=0.67) and the proportion of viable eggs in inseminated queens (ρ=0.48). The conventional live/dead assay of membrane permeability by dual fluorescent staining and a new test based on the leakage of the glycolytic enzyme glucose-phosphate-isomerase (GPI) from damaged cells were also correlated to the number of sperm reaching the spermatheca (ρ=0.54 and -0.61, respectively). We conclude that motility, live/dead-staining and the assay for GPI-leakage are valuable tools for the improvement of cryopreservation of honey bee semen.     

SIMEX variance component tests in generalized linear mixed measurement error models

In the analysis of clustered data with covariates measured with error, a problem of common interest is to test for correlation within clusters and heterogeneity across clusters. We examined this problem in the framework of generalized linear mixed measurement error models. We propose using the simulation extrapolation (SIMEX) method to construct a score test for the null hypothesis that all variance components are zero. A key feature of this SIMEX score test is that no assumptions need to be made regarding the distributions of the random effects and the unobserved covariates. We illustrate this test by analyzing Framingham heart disease data and evaluate its performance by simulation. We also propose individual SIMEX score tests for testing the variance components separately. Both tests can be easily implemented using existing statistical software.     

FiSSE: A simple nonparametric test for the effects of a binary character on lineage diversification rates

It is widely assumed that phenotypic traits can influence rates of speciation and extinction, and several statistical approaches have been used to test for correlations between character states and lineage diversification. Recent work suggests that model‐based tests of state‐dependent speciation and extinction are sensitive to model inadequacy and phylogenetic pseudoreplication. We describe a simple nonparametric statistical test (“FiSSE”) to assess the effects of a binary character on lineage diversification rates. The method involves computing a test statistic that compares the distributions of branch lengths for lineages with and without a character state of interest. The value of the test statistic is compared to a null distribution generated by simulating character histories on the observed phylogeny. Our tests show that FiSSE can reliably infer trait‐dependent speciation on phylogenies of several hundred tips. The method has low power to detect trait‐dependent extinction but can infer state‐dependent differences in speciation even when net diversification rates are constant. We assemble a range of macroevolutionary scenarios that are problematic for likelihood‐based methods, and we find that FiSSE does not show similarly elevated false positive rates. We suggest that nonparametric statistical approaches, such as FiSSE, provide an important complement to formal process‐based models for trait‐dependent diversification.     

Memory tests in healthy elderly: age factors and norms

The study aimed to assess the strength of age effects on both standard laboratory and ecological memory tests and the psychometric qualities of these tests. Furthermore, norm data are constructed. Memory performance was assessed in a random group of older adults (mean age 62 years, range 46-89) and a group of older adults having memory complaints who applied for memory training (mean age 63 years, range 45-85). Age effects were found on almost all memory tests, whether artificial laboratory or more ecological tests were used. Age effects remained generally present after correction for educational level. Retest reliabilities of the ecological memory tests did not differ systematically from those of standard laboratory tests. However, not all tests showed satisfactory retest reliabilities, this was even true for tests often used in clinical settings. For tests with retest reliabilities above r = .65 norms corrected for age and educational level were provided. The group of older adults having memory complaints performed on average better than the random group of older adults from the population. In the first group, higher performance thresholds should be employed in assessing whether memory performance is deviant.     

Interference of Manganese on neuroendocrinal system in exposed workers

The neurotoxic effects of Manganese (Mn) are well known, and are usually caused by high occupational exposure over long periods of time. The effects caused by low doses of this metal have been poorly evaluated. For this reason, we decided to evaluate neuroendocrinal tests in a group of 14 male workers employed in a ferrousmanganese foundry (exposed to Mn air concentrations within the TLV-TWA) and in 14 male control subjects, by analyzing FSH, LH, prolactin, and cortisol. The urinary Mn, prolactin, and cortisol levels were significantly higher in the worker’s group. The preliminary results of this research seem to show that for exposure below the TLV, Mn can somehow interfere with the neuroendocrine system. In order to confirm the existence of these effects and to verify their possible correlation with the dopaminergic control system, further studies are necessary.     

Effect of early and late exposure to estrual ewes on ram sexual performance classifications

This study was conducted to determine whether exposure of ram lambs to estrual ewes during their first autumn and again as adults just before serving capacity tests (SCT) affected the outcome of the sexual performance tests. Treatments were either early exposure of Polypay ram lambs (i.e., 7-8-mo-old rams with ewes for 17 d [n=30] or no early exposure [n=30]), and late exposure (i.e., 16-19-mo-old rams with estrual ewes for 3 d) or no exposure to estrual ewes in a 2x2 factorial arrangement. Three serving capacity tests were conducted immediately after the early exposure period for individual ram lambs that were exposed to ewes early. Three sham sexual performance tests (i.e., four ram lambs placed in test pens for 30-min without ewes) were conducted with ram lambs that were not exposed to ewes early. All rams were evaluated during nine 30-min serving capacity tests over a 2-mo period at 16-19 mo of age to determine sexual performance. Prior to serving capacity tests, one half of the rams from each early exposure treatment were exposed to estrual-induced ewes for 3 d. Specific sexual behaviors (e.g., sniffs, flehmens, foreleg kicks, vocalizations, mount attempts, mounts, and ejaculations) were recorded during serving capacity tests. Number of sniffs, flehmens, foreleg kicks, vocalizations, and mount attempts were summed without estimating the value of importance and analyzed as courtship behaviors. Sexual performance data were analyzed with Mixed model procedures for repeated measures. During serving capacity tests, the early exposed rams exhibited more courtships (40.3+/-8.0 versus 23.4+/-4.6; P<0.05; LSM+/-estimated SE), mounts (11.3+/-1.0 versus 7.7+/-0.9; P<0.01), and ejaculations (3.3+/-0.2 or 2.4+/-0.2; P<0.01) than rams not exposed to ewes as ram lambs, respectively. We conclude that early exposure of 7-8-mo-old ram lambs to estrual ewes improves sexual performance in serving capacity tests at 16-19 mo of age in most rams whereas, late exposure to estrual ewes for 3 d prior to serving capacity tests did not improve sexual performance scores.     

Neurobehavioral Dysfunction as a Possible Sentinel of Methylmercury Exposure

The main concern regarding methylmercury neurotoxicity relates to adverse effects on the brain during development. Many environmental chemicals may act as developmental neurotoxicants, but solid documentation from epidemiological studies exists only on methylmercury, lead, and polychlorinated biphenyls (PCBs). Neurobehavioral tests may reveal subtle dysfunctions, but the tests chosen must be valid and appropriate for the setting. In a prospective study in the Faroe Islands, the main neuropsychological functions affected by prenatal methylmercury exposure were attention, language and memory. Deficits in visuospatial function were mainly related to postnatal exposures. These associations were stable after adjustment for confounders and exclusion of the children with the highest exposures to methylmercury and PCBs. Tests with good psychometric properties were more likely to show an association with mercury exposure. Greater sensitivity was also seen with tests administered by specialized academic staff rather than a trained technician. Despite highly significant effects on nervous system function, the deficits were subtle, and mercury exposure explained only a small part of the variation. Available evidence suggests that neurotoxicity may have severe implications on public health, but current methods are not amenable to application as sentinels of adverse health effects in environmental health surveillance.     

Selenoprotein W as molecular target of methylmercury in human neuronal cells is down-regulated by GSH depletion

Methylmercury (MeHg) is well known as a neurotoxic chemical. However, little is mentioned about its neurotoxic mechanism or molecular target in human neuronal cells in particular. We show in this study that exposure of human neuronal cell line, SH-SY5Y, to MeHg dose- and time-dependently impairs viability and mRNA expression of selenoprotein W (SeW) with a significant difference, unlike other selenoenzymes such as, SeP, GPX4, 5DI, and 5'DI. Using real-time RT PCR, the influence of selenium (Se) and glutathione (GSH) on SeW expression was also investigated. While Se depletion caused a weakly reduced SeW mRNA levels, additional Se caused an increase of SeW mRNA levels. Although 2 mM GSH had induced a weak shift on SeW level, the expression of SeW mRNA was down-regulated in SH-SY5Y cells treated with 25 microM BSO, an inhibitor of GSH synthesis. To understand the relationship between a decrease of SeW expression and intracellular GSH and ROS, we measured the concentration of intracellular GSH and ROS in cells treated to 1.4 microM MeHg using fluorescence based assays. A positive correlation was found between SeW mRNA level and intracellular GSH but no significant correlation was observed between intracellular ROS and SeW mRNA level or intracellular GSH contents. Therefore, we suggest that SeW is the novel molecular target of MeHg in human neuronal cells and down-regulation of this selenoenzyme by MeHg is dependent not on generation of ROS but on depletion of GSH.     

Pyridine and other coal tar constituents as free radical-generating environmental neurotoxicants

Summary We have tested the hypothesis that chronic exposure to the principal constituents of the aqueous fraction of coal tar extracts can lead to the in vivo formation of substances which may produce neurological damage as the result of free radical generation and lipid peroxidation, these may be involved in the etiology of some neurological disorders. Artificial mixtures of the aqueous fraction of coal tar extracts were given in low concentrations to pigmented mice in their drinking water over a 3-month period. This resulted in significant increases in lipid peroxidation in the striatum, cerebellum and liver of the mice under test, the rank order being striatum > cerebellum > liver. These results are compatible with the possibility that coal tar emissions (as would be recovered or liberated in the burning, refining or beneficiation of coal) constitute a potential source of neurotoxicants with a predilection for damaging the nigrostriatal neuronal pathway. Our observations may thus have identified an important and hitherto unsuspected environmental source of neurotoxic chemicals, a possibility consistent with the proposed involvement of an environmental chemical factor in Parkinson's disease and perhaps in other neurological disorders.     

The Children's Treatment of Animals Questionnaire: A Rasch Analysis

ABSTRACT

The study of human–animal interactions is limited by a paucity of empirically validated measures of humane treatment of companion animals. The current study reports findings from a psychometric analysis of the Children's treatment of animals Questionnaire (CTAQ; thompson and Gullone 2003), an instrument that assesses children's humane interactions with nonhuman animals. Specifically, the current study extends what is known about the psychometric properties of the CTAQ by using traditional and item response theory analyses. The CTAQ was administered to a sample of 217 school-age children whose mothers were currently receiving residential or non-residential domestic violence services. Item-analysis, exploratory factor analysis, and parallel analysis were conducted to replicate previous psychometric evaluations of the CTAQ. Rasch analysis of the CTAQ was also conducted to provide a stringent test of unidimensionality and to identify potential invariance in item functioning across various demographic variables. The CTAQ showed adequate fit to the Rasch model; one modification, removal of item 5, was required. A Rasch principal components analysis of residuals indicated a single latent dimension among the remaining 12 items. Scale use was appropriate; Rasch-andrich thresholds increased with category values and no disordering of categories was evident. Examination of item-person maps indicated the sample was also well-targeted. Notably, evidence of differential item function was found across Spanish and English translations. Overall, findings indicate that the CTAQ is an appropriate unidimensional measure of children's humane treatment of animals. The measure is particularly well-suited for children ages 7 to 12 years who are at risk for exposure to and perpetration of animal cruelty. We recommend use of a 12-item version of the CTAQ to enhance the utility of the total score as a latent measure of children's humane treatment of companion animals.     

Analysis of multiple tumor data from a rodent carcinogenicity experiment

Rodent tumorigenicity experiments are conducted to determine the safety of substances for human exposure. The carcinogenicity of a substance is generally determined by statistical tests that compare the effects of treatment on the rate of tumor development at several body sites. The statistical analysis of such studies often includes hypothesis testing of the dose effect at each of the sites. However, the multiplicity of the significance tests may cause an excess overall false positive rate. In consideration of this problem, recent interest has focused on developing methods to test simultaneously for the treatment effect at multiple sites. In this paper, we propose a test that is based on the count of tumor-bearing sites. The test is appropriate regardless of tumor lethality or of treatment-related differences in the underlying mortality. Simulations are given which compare the performance of the proposed test to several other tests including a Bonferroni adjustment of site-specific tests, and the test is illustrated using the data from the large ED01 experiment.     

Dentine lead concentration as a predictor of neuropsychological functioning in inner-city children

Cumulative lead exposure in 193 inner-city Black children was evaluated by measuring lead concentrations in the primary dentine and the circumpulpal dentine of their deciduous teeth. The children’s dentine lead levels were comparable to those reported in population studies of low-income children living in inner-city areas. The results of neuropsychological tests administered at age 7 were available for children in the sample; we performed further neuropsychological assessments when their teeth were shed (10–14 yr of age). Higher dentine lead concentrations were associated with deficits in neuropsychological test performance at both testings, with performance abilities more strongly affected than verbal abilities. Moreover, the negative effects of lead on neuropsychological functioning were more evident when lead concentrations in circumpulpal dentine, rather than in primary dentine, was used as the index of lead exposure. Based on this finding, we suggest that circumpulpal dentine and primary dentine should be assayed separately in order to yield a more sensitive estimate of exposure to lead.