SEARCH · Search PubMed
Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
The notion that small, well planned clinical trials may not be worth undertaking is shown to arise from an overemphasis on just one way of interpreting P-values. Alternative forms of P and other interpretations are put forward. Attention is drawn to some aspects of the theory of hypothesis testing which seem less well known than they should be.
CRISCAT is a computer program for the analysis of grouped survival data with competing risks via weighted least squares methods. Competing risks adjustments are obtained from general matrix operations using many of the strategies employed in a previously developed program (GENCAT) for multivariate categorical data. CRISCAT computes survival rates at several time points for multiple causes of failure, where each rate is adjusted for other causes in the sense that failure due to thes other causes has been eliminated as a risk. The program can generate functions of the adjusted survival rates, to which asymptotic regression models may be fit. CRISCAT yields test statistics for hypotheses involving either these functions or estimated model parameters. Thus, this computational algorithm links competing risks theory to linear models methods for contingency table analysis and provides a unified approach to estimation and hypothesis testing of functions involving competing risks adjusted rates.
Both population-based and family-based designs are commonly used in genetic association studies to locate genes that underlie complex diseases. The simplest version of the family-based design--the transmission disequilibrium test--is well known, but the numerous extensions that broaden its scope and power are less widely appreciated. Family-based designs have unique advantages over population-based designs, as they are robust against population admixture and stratification, allow both linkage and association to be tested for and offer a solution to the problem of model building. Furthermore, the fact that family-based designs contain both within- and between-family information has substantial benefits in terms of multiple-hypothesis testing, especially in the context of whole-genome association studies.
Sinus rhythm (SR), ventricular tachycardia (VT) and ventricular fibrillation (VF) belong to different nonlinear physiological processes with different complexity. In this study, we present a novel, and computationally fast method to detect VT and VF, which utilizes a complexity measure suggested by Lempel and Ziv [1]. For a specific window length (i.e., the length of data segment to be analyzed), the method first generates a 0-1 string by comparing the raw electrocardiogram (ECG) data to a selected suitable threshold. The complexity measure can be obtained from the 0-1 string only using two simple operations, comparison and accumulation. When the window length is 7 s, the detection accuracy for each of SR, VT, and VF is 100% for a test set of 204 body surface records (34 SR, 85 monomorphic VT, and 85 VF). Compared with other conventional time- and frequency-domain methods, such as rate and irregularity, VF-filter leakage, and sequential hypothesis testing, the new algorithm is simple, computationally efficient, and well suited for real-time implementation in automatic external defibrillators (AED's).
The question of whether statistical significance testing should be used for the analysis of public health and epidemiologic data has received considerable attention in recent years. In this paper we have described some of the arguments for and against the use of hypothesis testing for the analysis of biomedical data. In addition, we have reviewed the literature from related fields, in particular sociology and psychology, in which similar discussions have taken place within the last 30 years. Many of the significance testing criticisms in these scientific fields have been raised in the more recent discussions taking place in the biomedical field. We present an example that emphasizes the use of both confidence interval estimation and significance testing. The example is particularly pertinent because it represents a more complex problem than has generally been discussed by critics of significance testing. Much of the discussion on this topic has focused on simple data analysis, such as the analysis of a 2 x 2 table or problems involving simple linear regression. Most epidemiologic data are far more complicated and warrant the use of both confidence interval estimation and significance testing for statistical analysis. Both of these techniques have no doubt been misused in the analysis of data. These misuses may have arisen from a lack of understanding of the role of statistical methods in data analysis and the choice of such methods for data analysis. If used prudently and judiciously, significance testing can help reduce the number of variables involved in a statistical analysis, thereby resulting in shorter confidence intervals for the models presented. Both significance testing and confidence interval estimation can serve and have served very useful functions for the analysis of public health and biomedical data.
One quality control test in the pharmaceutical industry is a test for uniformity of content of a batch prior to release of the batch to market. For batch acceptance by this or other quantitative tests of batch quality, one approach uses two-sided tolerance intervals of specified content. If the tolerance interval falls entirely within an acceptance interval, the batch is accepted. This has the form of a statistical hypothesis test. Once we recognize this approach as a statistical test, we can ask what sample size is required to be able to accept the batch with a desired power. The power for a single-stage design is a bivariate noncentral t probability and can be determined using previously published algorithms. Using standard methods for interim analyses, the approach is extended to multistage designs. Power and sample size for multistage designs are validated with simulations. We demonstrate it is possible to design one- and two-stage designs for batch acceptance with desired power and specified type I level.
Existing theory for inferences about linkage disequilibrium is restricted to a measure defined on gametic frequencies. Unless gametic frequencies are directly observable, they are inferred from genotypic frequencies under the assumption of random union of gametes. Primary emphasis in this paper is given to genotypic data, and disequilibrium coefficients are defined for all subsets of two or more of the four genes, two at each of two loci, carried by an individual. Linkage disequilibrium coefficients are defined for genes within and between gametes, and methods of estimating and testing these coefficients are given for gametic data. For genotypic data, when coupling and repulsion double heterozygotes cannot be distinguished. Burrows' composite measure of linkage disequilibrium is discussed. In particular, the estimate for this measure and hypothesis tests based on it are compared to the usual maximum likelihood estimate of gametic linkage disequilibrium, and corresponding likelihood ratio or contingency chi-square tests. General use of the composite measure, whether or not random union of gametes is an appropriate assumption, is recommended. Attention is given to small samples, where the non-normality of gene frequencies will have greatest effect on methods of inference based on normal theory. Even tools such as Fisher's z-transformation for the correlation of gene frequencies are found to perform quite satisfactorily.
Individual differences in acceptance patterns are evident as early as the child's first experiences with a particular food. To test hypothesis that the flavor of formula fed to infants modifies their acceptance of some foods, we conducted a within- and between-subjects design study in which two groups of 6- to 11-month-old infants were tested on two separate days. One group was currently feeding a milk-based formula whereas the other was feeding a protein hydrolysate formula, a particularly unpleasant tasting formula to adults that contains similar flavor notes (e.g., sulfur volatiles) with Brassica vegetables such as broccoli. In counterbalanced order, acceptance of pureed broccoli/cauliflower was determined during one test session and pureed carrots on the other. Although there were no group differences in the amount of carrots consumed, hydrolysate infants consumed significantly less broccoli/cauliflower relative to carrots when compared to those who were currently fed milk based formulas (F(1,72 df)=4.43; p=0.04). The mothers of hydrolysate infants were significantly more likely to report that their infants did not enjoy feeding the broccoli/cauliflower (54.2%) when compared to mothers of infants being fed milk-based formulas (28.0%; Chi-Square (1 df)=4.79; p=0.03). Such findings are consistent with prior research that demonstrated a sensory specific satiety following repeated exposure to a particular flavor in milk. We hypothesize that when infants are experiencing a flavor in milk or formula, in the short term, the preference that develops is specific to the context it is experienced in (e.g., milk). Over the longer term, the preference may generalize to other contexts such as solid foods. Hydrolysate infants were also significantly more likely to be judged by their mothers as being more active (F(1,69 df)=3.95; p=0.05) and hesitant (F(1,69 df)=6.55; p=0.01) when compared to those infants who were feeding milk-based formulas, a finding that further supports the hypothesis that mother-child dynamics surrounding early feeding impacts upon mothers' perception of their children's temperament.
This article reviews the misuse of statistical tests in psychotherapy research studies published in the Journal of Consulting and Clinical Psychology in the years 1967-1968, 1977-1978, and 1987-1988. It focuses on 3 major problems in statistical practice: inappropriate uses of null hypothesis tests and p values, neglect of effect size, and inflation of Type I error rate. The impressive frequency of these problems is documented, and changes in statistical practices over the past 3 decades are interpreted in light of trends in psychotherapy research. The article concludes with practical suggestions for rational application of statistical tests.
Science uses two types of data--one for generating hypotheses and a second for testing hypotheses. In clinical medicine the history and physical examination generate the data used for making hypotheses about what could possibly be wrong with the patient; but the various laboratory and imaging procedures are used for hypothesis-testing, ie, deciding which of the possible problems is actually the one bothering the patient. Although comparisons between physical examination maneuvers and imaging procedures may tell something about the precision of the former, in general such comparisons can tell little about the intrinsic worth of one versus the other, because these two different types of procedures generate data with two completely different types of usage.
Very few demographic surveys in developing countries have gathered information on household incomes or consumption expenditures. Researchers interested in living standards therefore have had little alternative but to rely on simple proxy indicators. The properties of these proxies have not been analyzed systematically. We ask what hypotheses can be tested using proxies, and compare these indicators with consumption expenditures per adult, our preferred measure of living standards. We find that the proxies employed in much demographic research are very weak predictors of consumption per adult. Nevertheless, hypothesis tests based on proxies are likely to be powerful enough to warrant consideration.
On the average, the circadian A of ML is larger in clinically healthy women at high vs. low RDBC. The original finding was made without numerical sampling on data from a relatively small group of women sampled systematically around the clock and the year. It was hence of further interest to see whether the risk-related difference can be corroborated by numerical resampling, i.e., by BS, a fairly general, usually computerintensive technique used for estimating the sampling distribution of an estimator or statistic (i.e., a quantity computed from the data). For BS of circannual parameters, one here assumes that 1) circadian As (obtained by the fit of a 24-hour cosine curve by linear least squares) are correlated for any given subject from season to season; and 2) deviations of each circadian A from the average are random. The method then estimates the characteristics of harmonic components, ordered by P-values obtained by linear least-squares analysis involving a test of the assumption of zero A for each special component. Results from BS validate differences in circadian A of ML between the low and high RDBC groups throughout the year. The use of BS serves to introduce inferential considerations into discriminant analysis and to test results obtained with very limited samples sizes, before embarking upon the actual labor-intensive, costly resampling required in circannual and circadian work. BS is a procedure recommended for broad routine use in non-parametric hypothesis testing and biomedical signal simulation. The re-evaluation of ML data by BS provides further inferential statistical validation of the finding as a first complement, not substitute, for additional sampling in order to assess a possibly important component of a test battery for breast cancer risk.
The hypothesis of "autoimmune testing" of mating partners assumes formation of individual system of perception on the basis of an immunologic principle, i.e. by lifetime selection of "direct" or "reverse casts" of own key antigens or signal molecules. Such system provides the coordinated change of signaling systems and system of their perception at formation of new adaptations that leads to automatic formation of reproductive isolation within the limited number of generations. Presence of the "friend-foe" recognition systems practically in all living organisms assumes formation of potential mating partners "autoimmune testing" mechanisms at the earliest evolution stages. In this article we analyze possible mechanisms of a "friend and foe" discrimination with MHC-proteins and their homologues in historical prospect--from bacteria to the lower Chordates.
Survival data with nonnegligible cure fractions are commonly encountered in clinical cancer clinical research. Recently, several authors (e.g. Kuk and Chen, Biometrika 79 (1992) 531; Maller and Zhou, Journal of Applied Probability, 30 (1993) 602; Peng and Dear, Biometrics, 56 (2000) 237; Sy and Taylor, Biometrics 56 (2000) 227) have proposed to use semiparametric cure models to analyze such data. Much of the existing work has been emphasized on cure detections and regression techniques. In contrast, this project focuses on the hypothesis testing in the presence of a cure fraction. Specifically, our interest lies in detecting whether there exists survival differences among noncured patients between treatment arms. For this purpose, we investigate the use of a modified Cramér-von Mises statistic for two-sample survival comparisons within the framework of cure models. Such a test has been studied by Tamura et al., (Statistics in Medicine 19, 2000, 2169) using bootstrap procedure. We will focus on developing asymptotic theory and convergent algorithms in this paper. We show that the limiting distributions of the Cramér-von Mises statistic under the null hypothesis can be represented by stochastic integrals and a weighted noncentral chi-squares. Both representations lead to concrete numerical schemes for computing the limiting distributions. The algorithms can be easily implemented for data analysis and significantly reduce computing time compared to the bootstrap approach. For illustrative purposes, we apply the proposed test to a published clinical trial.
MOTIVATION: Experimental design, hypothesis-testing and model-building in the current data-rich environment require the biologists' to collect, evaluate and integrate large amounts of information of many disparate kinds. Developing a unified framework for the representation and conceptual integration of biological data and processes is a major challenge in bioinformatics because of the variety of available data and the different levels of detail at which biological processes can be considered. RESULTS: We have developed the HyBrow (Hypothesis Browser) system as a prototype bioinformatics tool for designing hypotheses and evaluating them for consistency with existing knowledge. HyBrow consists of a modeling framework with the ability to accommodate diverse biological information sources, an event-based ontology for representing biological processes at different levels of detail, a database to query information in the ontology and programs to perform hypothesis design and evaluation. We demonstrate the HyBrow prototype using the galactose gene network in Saccharomyces cerevisiae as our test system, and evaluate alternative hypotheses for consistency with stored information. AVAILABILITY: www.hybrow.org
Hyperthyroidism has been reported to affect all of the major determinants of left ventricular performance in a manner that would augment ventricular shortening characteristics. The hypothesis tested in this study is that reduced afterload in conjunction with increased preload and heart rate, rather than augmented contractility, accounts for much of the increase in left ventricular performance noted previously in these patients. To investigate this hypothesis, 11 hyperthyroid patients were evaluated serially over 4 +/- 2 months. With therapy, serum total thyroxin (T4) decreased significantly (p less than 0.001). Ventricular hemodynamics were assessed by two-dimensional targeted M-mode echocardiograms and calibrated carotid pulse tracings. Ventricular preload was estimated by end-diastolic dimension, whereas afterload was measured as end-systolic wall stress. Overall left ventricular performance was quantitated by the extent and velocity of shortening, whereas myocardial work was assessed by ventricular systolic stress-length relations. With therapy, overall left ventricular performance declined (p less than 0.01). This change was associated with no change in end-diastolic dimension or end-systolic wall stress, and a 24% fall in heart rate (p less than 0.01). This latter finding has been shown previously to have no significant effect on left ventricular contractile state over the range of heart rates encountered in this study. In all cases, the end-systolic stress/rate-corrected shortening velocity relation fell with attainment of normal thyroid status, characteristic of a decline in contractility. There was a strong positive correlation between left ventricular contractility and serum thyroid hormone level (r = 0.83). In addition, ventricular minute work declined with therapy (p less than 0.01).(ABSTRACT TRUNCATED AT 250 WORDS)
The chi-square statistic has many scientific applications, including the evaluation of variance in counting data and the proper functioning of a radiation counting system. This paper provides a discussion of the fundamental aspects of the chi-square test using counting data. Practical applications of the chi-square statistic are discussed, including the estimation of extra-Poisson variance and dead time for a counting system. The consequences of passing or failing the chi-square test are discussed regarding the proper estimator for the population variance of the counting data. Example scenarios are used to provide insight into the applications of the chi-square statistic and the interpretation of values obtained in hypothesis testing.
A biomechanical analysis was performed of lifting before and after fatiguing the quadriceps muscles. The hypothesis tested was that when the quadriceps muscles were fatigued the lifter would change lifting technique from more of a squat (leg) lift to more of a stoop (back) lift to decrease the demand on the quadriceps muscles. The hypothesis was broadly supported, and three variables changed significantly with fatigue: trunk angular velocity, which increased, and knee moment integral and hip angles, which both decreased. These changes are all consistent with a change from more of a squat toward a stoop technique. The fact that the technique of lifting changes with quadriceps muscle fatigue underlines the importance of the physiologic condition of these muscles and suggests that rehabilitation of low-back-injured workers should include the quadriceps muscles. The amount of work performed should be controlled to avoid the development of local muscle fatigue and subsequent changes in performance.