Inverse sampling of controls in a matched case control study

A method of inverse sampling of controls in a matched case-control study is described in which, for each case, controls are sampled until a discordant set is achieved. For a binary exposure, inverse sampling is used to determine the number of controls for each case. When most individuals in a population have the same exposure, standard case-control sampling may result in many case-control sets being concordant with respect to exposure and thus uninformative in the conditional logistic analysis. The method using inverse control sampling is proposed as a solution to this problem in situations when it is practically feasible. In many circumstances, inverse control sampling is found to offer improved statistical efficiency relative to a comparable study with a fixed number of controls per case.     

Mechanisms of dusk orientation in white-throated sparrows (Zonotrichia albicollis): Clock-shift experiments

Summary Several species of night migrating birds, especially North American emberizines, exhibit markedly different orientation behaviour when tested in circular cages under clear skies at dusk as compared with tests performed after complete darkness. During the period between sunset and the first appearance of stars, birds tend to show high levels of well-oriented hopping; birds deprived of exposure to clear skies at dusk hop less and their activity is usually not oriented. There is evidence that visual cues available during the dusk period, but not later, are responsible for this difference, but details of the orientation mechanisms involved are unclear. We performed 3-h fast and slow clock shifts on white-throated sparrows (Zonotrichia albicollis) to address two questions concerning migratory orientation at dusk: (1) Is the better orientation of sparrows tested at dusk a function of the visual cues available at that time, or does it result from circadian changes in motivation?; and (2) Is the dusk orientation based on a time-compensated sun compass?Sparrows subjected to a 3-h slow clock shift were tested with controls on clear, moonless nights beginning immediately after lights-off in the clock shift room and thus about 3.5 h after local sunset. Individuals of both groups performed poorly oriented hopping typical of tests performed after complete darkness. The pooled data from each group were not significantly oriented. These results show that the visual cues available shortly after sunset, not temporal changes in the motivation of the birds, are responsible for the qualitative differences in orientation.Birds exposed to a 3-h fast clock shift were tested with controls on clear evenings between sunset and the first appearance of stars. Both groups showed well-oriented hopping. The mean direction of the pooled tests of controls was 325°, a typical spring orientation direction for this species. The mean direction of the pooled tests of the clock shifted birds (274°) was significantly different from that of controls and the 51° counterclockwise shift is consistent with that predicted by a time-compensated sun compass model.     

Two-sample rank tests for detecting changes that occur in a small proportion of the treated population

In the course of studying a biological phenomenon thought to be a precursor to chromosome breakage, researchers have found that treatments sometimes produce a higher proportion of "outliers" than do controls. Our examples pertain to smokers and patients undergoing chemotherapy, although the statistical methods developed here would apply to subjects exposed to any other health hazard. We formulate the problem in a nonparametric setting. Locally most powerful rank tests are obtained for mixture alternatives. In one instance, the approximate scores test has the simple form of counting the number of treatment responses above a combined sample percentile. Our test statistics are compared to the Wilcoxon and normal scores tests using empirical power studies and asymptotic efficiencies.     

Iron deficiency anemia and steady-state work performance at high altitude

Thirty-seven young adult male highland residents at 3,600-4,100 m in La Paz, Bolivia, performed short-duration cycle ergometry at 60, 80, and 100% of maximal voluntary O2 consumption (VO2max). Three groups of subjects representing the high-altitude population mean hemoglobin (Hb), the 10th percentile Hb, and below the 1st percentile were examined to test the hypothesis that the relationship of exercise performance to Hb concentration is similar to those relationships established at low altitude. Anemic individuals (n = 8) had 23% lower voluntary VO2max and 28% lower maximal work loads compared with controls (n = 17) or marginally anemic subjects (n = 12) although the relationship of VO2 to work load was similar. Anemic individuals maintained significantly higher arterial O2 partial pressures and Hb saturations during heavy exercise (90 +/- 0.5 vs. 85 +/- 0.6%) in conjunction with a greater heart rate up to maximal effort. A significantly decreased erythrocyte 2,3-diphosphoglycerate (2,3-DPG)-to-Hb molar ratio (0.70 +/- 0.04 vs. 1.12 +/- 0.06), suggestive of a left-shifted dissociation curve in anemics, is in contrast to the expected right-shifted curve. Moderate anemics were similar to controls. Anemic individuals did not differ in arterial lactate concentration from controls at absolute work loads; anemics had significantly lower arterial lactate concentrations at maximal effort than controls with no differences in the work load-to-lactate relationship. In conclusion, O2 transport during exercise at high altitude seems unaffected by the Hb concentrations as low as the 10th percentile of the population mean.(ABSTRACT TRUNCATED AT 250 WORDS)     

Testing hypotheses about interclass correlations from familial data

Testing problems concerning interclass correlations from familial data are considered in the case where the number of siblings varies among families. Under the assumption of multivariate normality, two test procedures are proposed for testing the hypothesis that an interclass correlation is equal to a specified value. To compare the properties of the tests, including a likelihood ratio test, Monte Carlo experiments are performed. Several test statistics are derived for testing whether two variables about a parent and child are uncorrelated. The proposed tests are compared with previous test procedures, using Monte Carlo simulation. A general procedure for finding confidence intervals for interclass correlations is also derived.     

A quantitative comparison of a percentile rule with a 2-fold rule for assessing mutagenicity in the Ames assay

Using a model based on the bivariate normal density function, this paper compares the effectiveness of two commonly employed decision rules for assessing mutagenicity in the standard Ames Salmonella assay. The 2-fold method, which considers a compound significantly mutagenic if its mean number of revertants per plate at any dose is equal to or greater than twice the mean number of revertants per plate in the concurrent control, may be a poor indicator of significant mutagenesis. In the percentile method, the frequency of induced mutations for the test compound is tested against the 95th percentile of the accumulated historical data for the spontaneous mutation frequency. As judged by the higher probability of declaring a compound mutagenic that elevates the reversion rate above background, the percentile rule is more reliable than the 2-fold method.     

A mixture-model approach for parallel testing for unequal variances

Testing for unequal variances is usually performed in order to check the validity of the assumptions that underlie standard tests for differences between means (the t-test and anova). However, existing methods for testing for unequal variances (Levene's test and Bartlett's test) are notoriously non-robust to normality assumptions, especially for small sample sizes. Moreover, although these methods were designed to deal with one hypothesis at a time, modern applications (such as to microarrays and fMRI experiments) often involve parallel testing over a large number of levels (genes or voxels). Moreover, in these settings a shift in variance may be biologically relevant, perhaps even more so than a change in the mean. This paper proposes a parsimonious model for parallel testing of the equal variance hypothesis. It is designed to work well when the number of tests is large; typically much larger than the sample sizes. The tests are implemented using an empirical Bayes estimation procedure which `borrows information' across levels. The method is shown to be quite robust to deviations from normality, and to substantially increase the power to detect differences in variance over the more traditional approaches even when the normality assumption is valid.     

Statistical tests of genetic association for case-control study designs

The central theme in case-control genetic association studies is to efficiently identify genetic markers associated with trait status. Powerful statistical methods are critical to accomplishing this goal. A popular method is the omnibus Pearson's chi-square test applied to genotype counts. To achieve increased power, tests based on an assumed trait model have been proposed. However, they are not robust to model misspecification. Much research has been carried out on enhancing robustness of such model-based tests. An analysis framework that tests the equality of allele frequency while allowing for different deviation from Hardy-Weinberg equilibrium (HWE) between cases and controls is proposed. The proposed method does not require specification of trait models nor HWE. It involves only 1 degree of freedom. The likelihood ratio statistic, score statistic, and Wald statistic associated with this framework are introduced. Their performance is evaluated by extensive computer simulation in comparison with existing methods.     

Effects of the pharmacologic manipulation of testosterone on cognitive functioning in women with polycystic ovary syndrome: a randomized, placebo-controlled treatment study

In a previous study, we found that women with polycystic ovary syndrome (PCOS), an endocrine disorder characterized by chronic hyperandrogenism, performed more poorly than healthy, matched controls on a number of neuropsychological tests, in particular tests of verbal fluency, verbal memory, manual dexterity, and visuospatial working memory. This randomized, placebo-controlled trial was undertaken to investigate whether pharmacologic manipulation of free testosterone (free T) levels in women with PCOS might affect their performance on cognitive tests. Nineteen women with PCOS completed a battery of neuropsychological tests before and after 3 months of treatment with either an anti-androgen (cyproterone acetate) plus estrogen or with a placebo. Hormone treatment of women with PCOS caused a significant reduction in their free T levels but did not affect performance on tests visuospatial ability, verbal memory, manual dexterity, or perceptual speed. Women treated with hormone therapy did, however, demonstrate an improvement in their performance on a test of verbal fluency compared to their pre-treatment scores. These findings suggest that changes in free T levels do not have a significant impact on cognitive performance in women with PCOS, although reductions in free T may be beneficial for verbal fluency.     

Biomonitoring of a worker population exposed to low antimony trioxide levels.

Antimony trioxide (Sb2O3) is used as a flame retardant in the textile industry. We carried out a study in a factory for the evaluation of antimony (Sb) occupational exposure and urinary levels in workers exposed to Sb2O3. Urinary levels and airborne Sb2O3 personal exposure values were very low when compared to international occupational standards (500 microg/m3, as Sb). The range of forty-two personal exposures was 0.01-0.55 microg Sb/m3, while twenty-four area samplings ranged from < 0.01 microg Sb/m3 to 1.45 microg Sb/m3. The mean urinary Sb levels at the beginning (n = 39) and end of the shift (n = 39) were 0.31 +/- 0.25 microg/L and 0.35 +/- 0.29 microg/L respectively, without any significant statistical difference. When the workers were divided into two subgroups according to "higher" and 'lower" exposure levels, a statistical difference (P < 0.001) was observed between the mean Sb urinary levels of the two subgroups during the workweek, both at the beginning and end of the shift. A statistical difference was also observed between the above mentioned subgroups and the controls (n = 15). No correlation was found between personal Sb2O3 exposure and the difference in urinary Sb levels at the beginning and end of the workshift on the day the flame retardant was utilized. This lack of correlation could be due to low airborne Sb2O3 levels and Sb dietary intake, estimated as 3 microg/day in UK, but not yet fully investigated in Italy. Any accidental occupational Sb per os exposure however low, could further enhance the lack of correlation.     

Chromosomal fragile site expression in lymphocytes from patients with schizophrenia

Schizophrenia is a common complex mental disorder. The lifetime prevalence of this disease is about 1% across different populations. The etiology is still unknown despite decades of intensive study. This report is aimed at studying the relationship between chromosomal fragile sites and the etiology of schizophrenia. Lymphocytes of 72 schizophrenic patients and 66 healthy controls were cultured in M medium, which is deficient in folic acid, and in medium RPMI1640 with distamycin A. G-banding was carried out on 100 metaphases of each individual. Fragile sites were characterized as specific chromosomal bands that exhibit nonrandom gaps or breaks. Culture in M medium resulted in significant differences in the total number of chromosomal lesions and the total number of cells with chromosomal lesions between patients and controls (P<0.001), while no difference was noted after exposure to distamycin A. In the case of M medium, 17 bands in both patients and controls were recognized as expressing fragile sites nonrandomly using a statistical method based on the relationship of the binomial and F distributions. Further analysis using Fisher’s exact test revealed a significant excess of expression of a rare fragile site at 2q11.2 among patients compared with controls (P<0.05). In the case of distamycin A induction, 13 bands were identified as having nonrandom expression of fragile sites using the same statistical method. A significant excess expression of a fragile site at 9q12 was identified among patients compared with controls by applying Fisher’s exact test (P<0.001). Thus, our data suggest that chromosomal bands 2q11.2 and 9q12 are interesting regions that may harbor important genes associated with schizophrenia. Received: 21 July 1998 / Accepted: 19 September 1998     

Effects of Obesity,Glucocorticoid, and Oral Contraceptive Therapy on Plasma Glucose and Blood Pyruvate Levels

The plasma glucose and blood pyruvate levels were determined after oral glucose tolerance test in six groups of women: non-obese and obese controls and in non-obese and obese women receiving glucocorticoid or oral contraceptive therapy. The mean fasting plasma glucose level was similar in all groups, but glucose tolerance was impaired in the obese controls, non-obese women on oral contraceptives or being treated with glucocorticoids, and appreciably impaired in the obese oral contraceptive and glucocorticoid groups compared with mean levels in non-obese subjects of the same groups. Obesity was associated with abnormally raised blood pyruvate levels in response to a glucose tolerance test in all groups. Striking similarities were observed between the responses of the plasma glucose and blood pyruvate levels to glucose tolerance tests in the obese control and non-obese oral contraceptive and non-obese glucocorticoid-treated groups. It is suggested that these abnormalities result from a common mechanism—namely, glucocorticoid excess.     

Electrophysiological and neuropsychological outcomes of severe obstructive sleep apnea: effects of hypoxemia on cognitive performance

Obstructive sleep apnea syndrome (OSAS) is a sleep disorder characterized with upper airway obstructions. Some studies showed cognitive and electrophysiological changes in patients with OSAS; however, contradictory results were also reported. The purpose of the present study was twofold: (1) to investigate cognitive changes in severe OSAS patients by using neuropsychological tests and electrophysiological methods together, (2) to investigate influence of hypoxemia levels on cognition. Fifty-four severe OSAS patients and 34 age-, gender- and education matched healthy subjects were participated. OSAS patients were further divided into two subgroups according to minimum oxygen saturation levels. All participants underwent a detailed neuropsychological test battery. A classical visual oddball task was used to elicit ERP P300 and mean P300 amplitudes were measured from F_z, C_z and P_z electrode sites. OSAS patients showed reduced mean P300 amplitudes up to 43–51% on all electrode sites compared to healthy controls. Subgroup analysis revealed significant differences in neuropsychological test scores between healthy controls and high hypoxemia OSAS group, as well as between low and high hypoxemia groups. Moreover, both low and high hypoxemia OSAS groups had lower P300 amplitudes compared with healthy controls. P300 amplitudes showed a gradual decline in parallel with increasing hypoxemia severity; however, the difference between high and low hypoxemia OSAS groups did not reach significance. Moderate correlations were found between sleep parameters, neuropsychological test scores and P300 amplitudes. These results suggest that electrophysiological measures could be better indicators of cognitive changes than neuropsychological tests in OSAS, particularly in mildly affected patients.     

The micronucleus assay as a biological dosimeter in hospital workers exposed to low doses of ionizing radiation

The health risk associated with low levels of ionizing radiation is still a matter of debate. A number of factors, such as non-target effects, adaptive responses and low-dose hypersensitivity, affect the long-term outcome of low-dose exposures. Cytogenetic bio-dosimetry provides a measure of the absorbed dose, taking into account the individual radiation sensitivity. The aim of the present study is to evaluate the value of the micronucleus (MN) test as a bio-dosimeter in hospital workers exposed to low doses of ionizing radiation. Blood samples were obtained from 30 subjects selected among workers exposed to X- and gamma-radiation, and 30 controls matched for sex, age and smoking from the same hospital. Micronucleus frequencies were analyzed by use of the cytokinesis-block method. The MN frequency was compared among the groups considering the confounding factors and the length of employment. No increase in the number of bi-nucleated cells with MN (BNMN), but a significant increase in the number of mono-nucleated cells with micronuclei (MOMN) was observed in exposed subjects compared with the controls. The relationship between MN frequency and accumulated dose (mSv) was evaluated. The length of employment did not affect the extent of MN frequency, but an increase of BNMN and MOMN cells was observed based on the accumulated radiation dose. Our study shows the sensitivity of the MN test in the detection of cytogenetic effects of cumulative exposure levels, suggesting the potential usefulness of this assay in providing a biological index in medical surveillance programs.     

Linear Rank Statistics for the Simple Tree Alternatives

The problem of comparing several treatments with a control is considered. It is formulated as a test of homogeneity with one-sided alternatives that at least one treatment is better than the control, while no treatment is worse than the control. A class of linear rank tests for the simple tree alternatives is proposed, assuming a location model. The most efficient test of the proposed class for detecting a specified pattern is derived. Two optimal tests for use in the case of vaguely specified tree alternatives are derived. In a simulation study the power of these two tests are compared with the power of Kruskal-Wallis test. Recommendations about the use of the proposed tests are made.     

Shift in diversification in sister-clade comparisons: a more powerful test

Tests of shift in diversification associated with key innovations or directional environmental change can be performed with sister-clade comparisons. This approach is attractive because it does not require detailed phylogenetic information. I propose a new likelihood ratio test based on fitting two models of diversification. I show how this test differs from a previous likelihood ratio test based on the geometric distribution. With simulations from a wide range of situations, I show that the new test performs much better than this test and the traditional test by Slowinski and Guyer. The proposed test performs at least as well as the species richness contrast test that has been proposed by several authors in four versions. A power analysis with low number of pairs of sister clades showed that the new test could detect a shift in diversification with five or less pairs of sister clades, whereas the diversity contrast test cannot detect any shift in this situation. The former appears as more powerful than the latter, and therefore is recommended when the number of pairs of sister clades is low (less than 10). All other tests should not be used as the present study showed they lack statistical power and robustness.     

A cytogenetic study of papaya workers exposed to ethylene dibromide

Ethylene dibromide (EDB) has been shown to be carcinogenic in animal studies and mutagenic in vitro. One cytogenetic study of workers exposed to low levels of EDB for short durations was negative. To test whether exposure to low levels of EDB over long periods caused cytogenetic changes, we have assessed the frequencies of sister-chromatid exchanges (SCE) and chromosomal aberrations (CA) in the peripheral blood lymphocytes of 60 men occupationally exposed to EDB. These men worked in papaya-packing plants where EDB was used to fumigate the fruit after harvest to kill fruit-fly larvae. 42 other men who worked at a nearby sugar mill served as controls. The average duration of exposure of the papaya workers was 5 years. 82 full shift personal breathing-zone air samples indicated that the papaya workers were exposed to a geometric mean of 88 ppb of EDB, as an 8-h time weighted average (TWA). Peaks up to 262 ppb were measured. The proposed OSHA 8-h TWA for EDB is 100 ppb, while NIOSH recommends 45 ppb. No differences in SCE levels were found between exposed and nonexposed workers. No differences were found in the total CA frequency between exposed and nonexposed workers. SCE levels were significantly increased in men who smoked cigarettes (p = 0.0001) and in men who smoked marijuana (p = 0.01). CA levels showed a significant increasing trend with age (p = 0.03).     

Model-specific tests on variance heterogeneity for detection of potentially interacting genetic loci

ABSTRACT: BACKGROUND: Trait variances among genotype groups at a locus are expected to differ in the presence of an interaction between this locus and another locus or environment. A simple maximum test on variance heterogeneity can thus be used to identify potentially interacting single nucleotide polymorphisms (SNPs). RESULTS: We propose a multiple contrast test for variance heterogeneity that compares the mean of Levene residuals for each genotype group with their average as an alternative to a global Levene test. We applied this test to a Bogalusa Heart Study dataset to screen for potentially interacting SNPs across the whole genome that influence a number of quantitative traits. A user-friendly implementation of this method is available in the R statistical software package multcomp. CONCLUSIONS: We show that the proposed multiple contrast test of model-specific variance heterogeneity can be used to test for potential interactions between SNPs and unknown alleles, loci or covariates and provide valuable additional information compared with traditional tests. Although the test is statistically valid for severely unbalanced designs, care is needed in interpreting the results at loci with low allele frequencies.     

Glycophorin A somatic cell mutations in a population living in the proximity of the Semipalatinsk nuclear test site

The glycophorin A (GPA) somatic mutation assay was performed to evaluate the magnitude of exposure to ionizing radiation among the human population living in the vicinity of the Semipalatinsk nuclear test site in Kazakhstan. All together, 113 blood samples were analyzed from three generations of people living in villages that were under the trail of the radioactive cloud from the first Soviet surface nuclear test performed in August 1949 and from later tests. The oldest generation (P0) lived in the area at the time of testing, whereas the younger generations (F1, F2) were exposed to smaller doses from the residual fallout and later tests. The GPA assay did not reveal significant differences in the variant cell frequencies for all subjects selected from the Semipalatinsk area compared with 74 matched controls living in a noncontaminated area. However, a significant increase (P < 0.05) in the mean allele-loss ON variant frequency was observed among the exposed P0 generation (12 x 10(-6)) in comparison to controls (7 x 10(-6)). Considering the sensitivity of the GPA assay, the results suggest that the mean dose to the P0 generation of the affected villages was relatively low, a finding which is in accordance to the conclusions obtained from other biological assays performed on the same population.     

Bivariate random‐effects meta‐analysis models for diagnostic test accuracy studies using arcsine‐based transformations

Diagnostic or screening tests are widely used in medical fields to classify patients according to their disease status. Several statistical models for meta‐analysis of diagnostic test accuracy studies have been developed to synthesize test sensitivity and specificity of a diagnostic test of interest. Because of the correlation between test sensitivity and specificity, modeling the two measures using a bivariate model is recommended. In this paper, we extend the current standard bivariate linear mixed model (LMM) by proposing two variance‐stabilizing transformations: the arcsine square root and the Freeman–Tukey double arcsine transformation. We compared the performance of the proposed methods with the standard method through simulations using several performance measures. The simulation results showed that our proposed methods performed better than the standard LMM in terms of bias, root mean square error, and coverage probability in most of the scenarios, even when data were generated assuming the standard LMM. We also illustrated the methods using two real data sets.