Sample size planning for rank-based multiple contrast tests.

Rank methods are well-established tools for comparing two or multiple (independent) groups. Statistical planning methods for the computing the required sample size(s) to detect a specific alternative with predefined power are lacking. In the present paper, we develop numerical algorithms for sample size planning of pseudo-rank-based multiple contrast tests. We discuss the treatment effects and different ways to approximate variance parameters within the estimation scheme. We further compare pairwise with global rank methods in detail. Extensive simulation studies show that the sample size estimators are accurate. A real data example illustrates the application of the methods.

Generalized Pairwise Comparisons to Assess Treatment Effects: JACC Review Topic of the Week.

A time-to-first-event composite endpoint analysis has well-known shortcomings in evaluating a treatment effect in cardiovascular clinical trials. It does not fully describe the clinical benefit of therapy because the severity of the events, events repeated over time, and clinically relevant nonsurvival outcomes cannot be considered. The generalized pairwise comparisons (GPC) method adds flexibility in defining the primary endpoint by including any number and type of outcomes that best capture the clinical benefit of a therapy as compared with standard of care. Clinically important outcomes, including bleeding severity, number of interventions, and quality of life, can easily be integrated in a single analysis. The treatment effect in GPC can be expressed by the net treatment benefit, the success odds, or the win ratio. This review provides guidance on the use of GPC and the choice of treatment effect measures for the analysis and reporting of cardiovascular trials.

Win odds: An adaptation of the win ratio to include ties.

The win ratio, a recently proposed measure for comparing the benefit of two treatment groups, allows ties in the data but ignores ties in the inference. In this article, we highlight some difficulties that this can lead to, and we propose to focus on the win odds instead, a modification of the win ratio which takes ties into account. We construct hypothesis tests and confidence intervals for the win odds, and we investigate their properties through simulations and in a case study. We conclude that the win odds should be preferred over the win ratio.

Optimal sample size planning for the Wilcoxon-Mann-Whitney test.

There are many different proposed procedures for sample size planning for the Wilcoxon-Mann-Whitney test at given type-I and type-II error rates α and ß, respectively. Most methods assume very specific models or types of data to simplify calculations (eg, ordered categorical or metric data, location shift alternatives, etc). We present a unified approach that covers metric data with and without ties, count data, ordered categorical data, and even dichotomous data. For that, we calculate the unknown theoretical quantities such as the variances under the null and relevant alternative hypothesis by considering the following "synthetic data" approach. We evaluate data whose empirical distribution functions match the theoretical distribution functions involved in the computations of the unknown theoretical quantities. Then, well-known relations for the ranks of the data are used for the calculations. In addition to computing the necessary sample size N for a fixed allocation proportion t = n1 /N, where n1 is the sample size in the first group and N = n1 + n2 is the total sample size, we provide an interval for the optimal allocation rate t, which minimizes the total sample size N. It turns out that, for certain distributions, a balanced design is optimal. We give a characterization of such distributions. Furthermore, we show that the optimal choice of t depends on the ratio of the two variances, which determine the variance of the Wilcoxon-Mann-Whitney statistic under the alternative. This is different from an optimal sample size allocation in case of the normal distribution model.

Canine distemper virus in the Serengeti ecosystem: molecular adaptation to different carnivore species.

Was the 1993/1994 fatal canine distemper virus (CDV) epidemic in lions and spotted hyaenas in the Serengeti ecosystem caused by the recent spillover of a virulent domestic dog strain or one well adapted to these noncanids? We examine this question using sequence data from 13 'Serengeti' strains including five complete genomes obtained between 1993 and 2011. Phylogenetic and haplotype network analyses reveal that strains from noncanids during the epidemic were more closely related to each other than to those from domestic or wild canids. All noncanid 'Serengeti' strains during the epidemic encoded: (1) one novel substitution G134S in the CDV-V protein; and (2) the rare amino acid combination 519I/549H at two sites under positive selection in the region of the CDV-H protein that binds to SLAM (CD 150) host cell receptors. Worldwide, only a few noncanid strains in the America II lineage encode CDV-H 519I/549H. All canid 'Serengeti' strains during the epidemic coded CDV-V 134G, and CDV-H 519R/549Y, or 519R/549H. A functional assay of cell entry revealed the highest performance by CDV-H proteins encoding 519I/549H in cells expressing lion SLAM receptors, and the highest performance by proteins encoding 519R/549Y, typical of dog strains worldwide, in cells expressing dog SLAM receptors. Our findings are consistent with an epidemic in lions and hyaenas caused by CDV variants better adapted to noncanids than canids, but not with the recent spillover of a dog strain. Our study reveals a greater complexity of CDV molecular epidemiology in multihost environments than previously thought.

A wild bootstrap approach for the selection of biomarkers in early diagnostic trials.

BACKGROUND: In early diagnostic trials, particularly in biomarker studies, the aim is often to select diagnostic tests among several methods. In case of metric, discrete, or even ordered categorical data, the area under the receiver operating characteristic (ROC) curve (denoted by AUC) is an appropriate overall accuracy measure for the selection, because the AUC is independent of cut-off points. METHODS: For selection of biomarkers the individual AUC's are compared with a pre-defined threshold. To keep the overall coverage probability or the multiple type-I error rate, simultaneous confidence intervals and multiple contrast tests are considered. We propose a purely nonparametric approach for the estimation of the AUC's with the corresponding confidence intervals and statistical tests. This approach uses the correlation among the statistics to account for multiplicity. For small sample sizes, a Wild-Bootstrap approach is presented. It is shown that the corresponding intervals and tests are asymptotically exact. RESULTS: Extensive simulation studies indicate that the derived Wild-Bootstrap approach keeps and exploits the nominal type-I error at best, even for high accuracies and in case of small samples sizes. The strength of the correlation, the type of covariance structure, a skewed distribution, and also a moderate imbalanced case-control ratio do not have any impact on the behavior of the approach. A real data set illustrates the application of the proposed methods. CONCLUSION: We recommend the new Wild Bootstrap approach for the selection of biomarkers in early diagnostic trials, especially for high accuracies and small samples sizes.

Evaluation of medical research performance--position paper of the Association of the Scientific Medical Societies in Germany (AWMF).

OBJECTIVE: The evaluation of medical research performance is a key prerequisite for the systematic advancement of medical faculties, research foci, academic departments, and individual scientists' careers. However, it is often based on vaguely defined aims and questionable methods and can thereby lead to unwanted regulatory effects. The current paper aims at defining the position of German academic medicine toward the aims, methods, and consequences of its evaluation. METHODS: During the Berlin Forum of the Association of the Scientific Medical Societies in Germany (AWMF) held on 18 October 2013, international experts presented data on methods for evaluating medical research performance. Subsequent discussions among representatives of relevant scientific organizations and within three ad-hoc writing groups led to a first draft of this article. Further discussions within the AWMF Committee for Evaluation of Performance in Research and Teaching and the AWMF Executive Board resulted in the final consented version presented here. RESULTS: The AWMF recommends modifications to the current system of evaluating medical research performance. Evaluations should follow clearly defined and communicated aims and consist of both summative and formative components. Informed peer reviews are valuable but feasible in longer time intervals only. They can be complemented by objective indicators. However, the Journal Impact Factor is not an appropriate measure for evaluating individual publications or their authors. The scientific "impact" rather requires multidimensional evaluation. Indicators of potential relevance in this context may include, e.g., normalized citation rates of scientific publications, other forms of reception by the scientific community and the public, and activities in scientific organizations, research synthesis and science communication. In addition, differentiated recommendations are made for evaluating the acquisition of third-party funds and the promotion of junior scientists. CONCLUSIONS: With the explicit recommendations presented in the current position paper, the AWMF suggests enhancements to the practice of evaluating medical research performance by faculties, ministries and research funding organizations.

Are multiple contrast tests superior to the ANOVA?

Multiple contrast tests can be used to test arbitrary linear hypotheses by providing local and global test decisions as well as simultaneous confidence intervals. The ANOVA-F-test on the contrary can be used to test the global null hypothesis of no treatment effect. Thus, multiple contrast tests provide more information than the analysis of variance (ANOVA) by offering which levels cause the significance. We compare the exact powers of the ANOVA-F-test and multiple contrast tests to reject the global null hypothesis. Hereby, we compute their least favorable configurations (LFCs). It turns out that both procedures have the same LFCs under certain conditions. Exact power investigations show that their powers are equal to detect their LFCs.

Analysis of Predictive Values Based on Individual Risk Factors in Multi-Modality Trials.

The accuracy of diagnostic tests with binary end-points is most frequently measured by sensitivity and specificity. However, from the clinical perspective, the main purpose of a diagnostic agent is to assess the probability of a patient actually being diseased and hence predictive values are more suitable here. As predictive values depend on the pre-test probability of disease, we provide a method to take risk factors influencing the patient's prior probability of disease into account, when calculating predictive values. Furthermore, approaches to assess confidence intervals and a methodology to compare predictive values by statistical tests are presented. Hereby the methods can be used to analyze predictive values of factorial diagnostic trials, such as multi-modality, multi-reader-trials. We further performed a simulation study assessing length and coverage probability for different types of confidence intervals, and we present the R-Package facROC that can be used to analyze predictive values in factorial diagnostic trials in particular. The methods are applied to a study evaluating CT-angiography as a noninvasive alternative to coronary angiography for diagnosing coronary artery disease. Hereby the patients' symptoms are considered as risk factors influencing the respective predictive values.

Antagonistic pleiotropy and fitness trade-offs reveal specialist and generalist traits in strains of canine distemper virus.

Theoretically, homogeneous environments favor the evolution of specialists whereas heterogeneous environments favor generalists. Canine distemper is a multi-host carnivore disease caused by canine distemper virus (CDV). The described cell receptor of CDV is SLAM (CD150). Attachment of CDV hemagglutinin protein (CDV-H) to this receptor facilitates fusion and virus entry in cooperation with the fusion protein (CDV-F). We investigated whether CDV strains co-evolved in the large, homogeneous domestic dog population exhibited specialist traits, and strains adapted to the heterogeneous environment of smaller populations of different carnivores exhibited generalist traits. Comparison of amino acid sequences of the SLAM binding region revealed higher similarity between sequences from Canidae species than to sequences from other carnivore families. Using an in vitro assay, we quantified syncytia formation mediated by CDV-H proteins from dog and non-dog CDV strains in cells expressing dog, lion or cat SLAM. CDV-H proteins from dog strains produced significantly higher values with cells expressing dog SLAM than with cells expressing lion or cat SLAM. CDV-H proteins from strains of non-dog species produced similar values in all three cell types, but lower values in cells expressing dog SLAM than the values obtained for CDV-H proteins from dog strains. By experimentally changing one amino acid (Y549H) in the CDV-H protein of one dog strain we decreased expression of specialist traits and increased expression of generalist traits, thereby confirming its functional importance. A virus titer assay demonstrated that dog strains produced higher titers in cells expressing dog SLAM than cells expressing SLAM of non-dog hosts, which suggested possible fitness benefits of specialization post-cell entry. We provide in vitro evidence for the expression of specialist and generalist traits by CDV strains, and fitness trade-offs across carnivore host environments caused by antagonistic pleiotropy. These findings extend knowledge on CDV molecular epidemiology of particular relevance to wild carnivores.

Estimation of Box's ε for low- and high-dimensional repeated measures designs with unequal covariance matrices.

We present new inference methods for the analysis of low- and high-dimensional repeated measures data from two-sample designs that may be unbalanced, the number of repeated measures per subject may be larger than the number of subjects, covariance matrices are not assumed to be spherical, and they can differ between the two samples. In comparison, we demonstrate how crucial it is for the popular Huynh-Feldt (HF) method to make the restrictive and often unrealistic or unjustifiable assumption of equal covariance matrices. The new method is shown to maintain desired α-levels better than the well-known HF correction, as demonstrated in several simulation studies. The proposed test gains power when the number of repeated measures is increased in a manner that is consistent with the alternative. Thus, even increasing the number of measurements on the same subject may lead to an increase in power. Application of the new method is illustrated in detail, using two different real data sets. In one of them, the number of repeated measures per subject is smaller than the sample size, while in the other one, it is larger.

Sibling rivalry: training effects, emergence of dominance and incomplete control.

Within-brood or -litter dominance provides fitness-related benefits if dominant siblings selfishly skew access to food provided by parents in their favour. Models of facultative siblicide assume that dominants exert complete control over their subordinate sibling's access to food and that control is maintained, irrespective of the subordinate's hunger level. By contrast, a recent functional hypothesis suggests that subordinates should contest access to food when the cost of not doing so is high. Here, we show that within spotted hyena (Crocuta crocuta) twin litters, dominants most effectively skew access to maternal milk in their favour when their aggression prompts a highly submissive response. When hungry, subordinates were less submissive in response to aggression, thereby decreasing lost suckling time and increasing suckling time lost by dominants. In a species where adult females socially dominate adult males, juvenile females were more often dominant than males in mixed-sex litters, and subordinate sisters used more effective counter-tactics against dominant brothers than subordinate brothers against dominant sisters. Our results provide, to our knowledge, the first evidence in a mammal that dominant offspring in twin litters do not exert complete control over their sibling's access to resources (milk), and that sibling dominance relationships are influenced by sibling sex and training effects.

Comparison of global tests for functional gene sets in two-group designs and selection of potentially effect-causing genes.

MOTIVATION: An important object in the analysis of high-throughput genomic data is to find an association between the expression profile of functional gene sets and the different levels of a group response. Instead of multiple testing procedures which focus on single genes, global tests are usually used to detect a group effect in an entire gene set. In a simulation study, we compare the power and computation times of four different approaches for global testing. The applicability of one of these methods to gene expression data is demonstrated for the first time. In addition, we propose an algorithm for the detection of those genes which might be responsible for a group effect. RESULTS: We could detect that the power of three of the approaches is comparable in many settings but considerable differences were detected in the computation times. Our proposed gene selection algorithm was able to detect potentially effect-causing genes in artificial sets with high power when many genes were altered with a small effect, while classical multiple testing was more powerful when few genes were altered with a large effect. AVAILABILITY: An R-package called 'RepeatedHighDim' which implements our new global test procedures is made available from http://cran.r-project.org/.

The endogenous danger signals HSP70 and MICA cooperate in the activation of cytotoxic effector functions of NK cells.

Although natural killer (NK) cells are often described as first line defence against infected or malignant cells which act without the need of prior activation, it is known now that the NK cell activity is tightly regulated by other cells and soluble factors. We show here that the stress-inducible heat shock protein (HSP) 70 activates human NK cells to kill target cells expressing major histocompatibility complex class I chain-related molecule A (MICA) in a natural killer group 2 member D (NKG2D-) dependent manner. The HSP70-derived peptide TKD (TKDNNLLGRFELSG) was able to replace the full-length HSP70 and to exert the same function. Interestingly, the expression of the cytotoxic effector protease granzyme B in NK cells was increased after TKD stimulation. When MICA and MICB expression was induced in human tumour cells by a histone deacetylase inhibitor and NK cells were activated by HSP70 or TKD, both treatments jointly improved the killing of the tumour cells. Thus, the synergistic activity of two stress-inducible immunological danger signals, HSP70 and MICA/B, leads to activation and enhanced cytotoxicity of human NK cells against tumour cells.

Robustness considerations in selecting efficient two-color microarray designs.

The main goal of microarray experiments is to select a small subset of genes that are differentially expressed among competing mRNA samples. For a given set of such mRNA samples, it is possible to consider a number of two-color cDNA microarray designs with a fixed number of arrays. Appropriate criteria can be used to select an efficient design from such a set of alternative experimental designs. In practice, however, microarray expression data often contain missing observations and the most efficient design (with complete observations) for a specific setup may not be efficient in the presence of missing observations. In this article, we propose two criteria to address the robustness of microarray designs against missing observations. We demonstrate the simultaneous use of efficiency and robustness criteria to select good microarray designs for both one-factor and multi-factor experiments.

Digital image color analysis compared to direct dental CIE colorimeter assessment under different ambient conditions.

PURPOSES: To evaluate the concordance and repeatability of two in vivo methods for dental color assessment and to clarify the influence of different ambient light conditions and subject's head position on the assessed color variables. METHODS: Color assessments were performed by two examiners on 16 arbitrarily selected subjects under two different, standardized conditions of illumination and at two different standardized head angulations. CIE (L*a*b*) data for upper and lower central incisors were recorded in two different ways: (1) by an intra-oral contact dental colorimeter and (2) by processing digital images for performing color calculation using Adobe Photoshop software. The influence of the different ambient conditions on both methods, as well as the concordance of measurements was analyzed statistically using several mixed linear models. RESULTS: Ambient light as a single factor had no significant influence on maxillary L*, a* and b* values, but it did have an effect on mandible assessments. Head angulation variation resulted in significant L* value differences using the photo method. The operator had a significant influence on values a* and b* for the photo method and on a* values for the colorimeter method. In fully lit ambient condition, the operator had a significant influence on the segregated L*, a*, and b* values. With dimmed lights, head angulation became significant, but not the operator. Evaluation of segregated L* values was error prone in both methods. Comparing both methods, deltaE values did not exceed 2.85 units, indicating that color differences between methods and recorded under varying ambient conditions were well below the sensitivity of the naked eye.

Individual versus group oral hygiene instruction for adults.

PURPOSE: The objective of this study was to assess the practicability and effectiveness of group oral hygiene instruction for adults in comparison with individual oral hygiene instruction. MATERIALS AND METHODS: A total of 104 male subjects aged 18 to 54 (mean age: 29.7) years were randomly assigned to one of four groups: group A was given individual oral hygiene instruction; group B was given group oral hygiene instruction; group C was given a combination of individual and group oral hygiene instructions; and group D received no oral hygiene instruction (the control group). The success of each form of instruction was evaluated on the basis of four parameters: (1) the Quigley-Hein plaque index (QHI), (2) the approximal plaque index (API), (3) a modified sulcus bleeding index (SBI) and (4) the community periodontal index of treatment needs (CPITN). All participants had professional tooth cleaning at the end of the baseline examination. The final examinations were conducted 13 weeks later. RESULTS: All subjects showed a poor oral health status at the beginning of the study, and the mean QHI score was 2.2. In addition, 92% of all subjects had an API score of more than 70%. Moderate-to-severe gingival inflammation (modified SBI) was observed in 67.3% of the subjects. CPITN scores of 2 or 3 were calculated for 82% of all sextants. At the end of the study, all groups showed a significant improvement in their oral health status and periodontal parameters (P < 0.0001). The majority of the subjects achieved an API score between 25% and 70%, and they had a mean QHI score of 1.2. A CPITN score of 0 or 1 was recorded for most sextants (62%). There was no significant difference between the various groups. CONCLUSIONS: This study demonstrated that group oral hygiene instruction and conventional individual instruction have similar beneficial effects in adults.

Decontamination efficacy of erbium:yttrium-aluminium-garnet and diode laser light on oral Candida albicans isolates of a 5-day in vitro biofilm model.

The different forms of superficial and systemic candidiasis are often associated with biofilm formation on surfaces of host tissues or medical devices. The biofilm formation of Candida spp., in general, necessitates significantly increased amounts of antifungal agents for therapy. Often the therapeutic effect is doubtful. A 5-day biofilm model with oral Candida isolates was established according to Chandra et al. (J Dent Res 80:903-908, 2001) on glass and titanium surfaces and was modified by Sennhenn-Kirchner et al. (Z Zahnärztl Implantol 3:45-51, 2007) to investigate different aspects unanswered in the field of dentistry. In this model, the efficacy of erbium:yttrium-aluminium-garnet (Er:YAG) light (2940 nm, 100 mJ, 10 Hz, 300 micros pulsed mode applied for 80 s) and diode laser light (810 nm, 1 W, continuous wave mode applied for 20 s with four repetitions after 30 s pauses each) was evaluated and compared to untreated controls. The photometric evaluation of the samples was completed by observations on morphological changes of yeast cells grown in the biofilm. Compared to the untreated controls Candida cells grown in mature in vitro biofilms were significantly reduced by both wavelengths investigated. Comparison between the different methods of laser treatment additionally revealed a significantly greater effect of the Er:YAG over the diode laser. Scanning electron microscopy findings proved that the diode laser light was effective in direct contact mode. In contrast, in the areas without direct contact, the fungal cells were left almost unchanged. The Er:YAG laser damaged the fungal cells to a great extent wherever it was applied.

[Oral health in young adults in Germany--a comparison between women and men of the German army]. / Mundgesundheit von jungen Frauen und Männern in Deutschland. Ein Vergleich von Angehörigen der Bundeswehr.

The objective of this cross-sectional study was to compare oral health in young German women and men. We paired female and male soldiers of the German armed force (Bundeswehr) on the basis of age, school qualification and origin. All participants underwent a clinical examination. Oral health was assessed using the decayed, missing and filled surfaces index (DMF-S), an oral hygiene index (QHI), and a periodontal index (CPITN). In addition, we collected data on the subjects' smoking habits. The study included 90 women (mean age: 21.7 years) and 90 men (mean age: 21.4 years). The mean DMF-S was 20.7 +/- 13.7 for women and 21.0 +/- 19.8 for men (p>0.05). In the female group, dental restorations were present in 71% of the teeth affected by caries. In the male group, restorations were present in 61.2% (p=0.04). The female group showed a less marked polarisation of caries than the male group (p=0.03). Oral hygiene (QHI) was poor in the entire study population. Women (QHI: 2.16 +/- 0.6), however, exhibited significantly better oral hygiene than men (QHI: 2.53 +/- 0.5) (p= 0.01). Deeper probing depths (CPITN scores of 3 and 4) were less frequently seen in females than in males (p= 0.04). In both groups more than half of the participants were smokers (56.7%) since at least four years with more than ten cigarettes per day. Smoking had a significant influence on periodontal health (p=0.02). Men showed poorer oral health than women. The study emphasizes that the proportion of young adults requiring curative dental treatment remains high in Germany. It also underlines the need for intensive preventive care including cessation of smoking.

A parametric analysis of ordinal quality-of-life data can lead to erroneous results.

OBJECTIVE: Measurements from health-related quality-of-life (HRQoL) studies, although usually of an ordered categorical nature, are typically treated as continuous variables, allowing the calculation of mean values and the administration of parametric statistics, such as t-tests. We investigated whether parametric, compared to nonparametric, analyses of ordered categorical data may lead to different conclusions. STUDY DESIGN AND SETTING: HRQoL data were obtained from patients with a diagnosis of asthma (n=192) and chronic obstructive pulmonary disease (COPD; n=88) at two time points. The impact of the group factor (asthma vs. COPD) and the time factor (t1 vs. t2) on HRQoL was analyzed with a metric approach (repeated measures ANOVA) and two ordinal approaches (each with a nonparametric repeated measures ANOVA). RESULTS: Using the metric approach, a significant effect of "group" (P=0.0061) and "time" (P=0.0049) on HRQoL was found. The first ordinal approach (ranked total score) still showed a significant effect for "group" (P=0.0033) with a worse HRQoL for patients suffering from COPD. In the second approach (ranks for each HRQoL item and summed ranks), there were no significant effects. CONCLUSION: Applying simple parametric methods to ordered categorical HRQoL scores led to different results from those obtained with nonparametric methods. In these cases, an ordinal approach will prevent inappropriate conclusions.