Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “hypothesis testing”

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.

At least 289 records · Page 16Linked to original sources
PubMed2002

Randomized, placebo-controlled, parallel group versus crossover study designs for the study of dementia in Parkinson's disease.

In studies of dementia, crossover designs are controversial, reflecting concerns about temporal stability of disease, confounding of treatment effects with period by treatment interactions and/or carryover effects. Carryover effects are differences in the lingering effect of treatments (placebo) into subsequent periods. In the context of a trial to study the effect of donepezil on dementia in patients with Parkinson's disease, we examine two-sequence crossover studies with two or four periods, and a four-sequence design with two periods. We quantify bias in estimated treatment effects due to carryover effects and explore the use of biased estimators in hypothesis testing. For hypothesis testing, type I error rates are valid if (1) repeated administration of treatment alters the outcome only for effective treatments and (2) carryover effects due to placebo following treatment periods are nonzero only for effective treatments. For crossover and parallel group designs, sample sizes are adjusted for reduced statistical power due to carryover effects and temporal changes in variance. For the proposed clinical study, we estimate that a single-period parallel group design with baselines would require 104 patients and take about 23 months to complete. A two-sequence, four-period parallel group design with baselines would require about 80 patients and about 20 months to complete. We conservatively assume a carryover effect of 50% of the treatment effect for a two-sequence four-period crossover design. The estimated treatment effect for this model may underestimate the true treatment effect by up to 13%. The sample size/study length requirements are 28 patients or 12.4 months, respectively, a substantial saving over either parallel group design. The cost of allowing for carryover in the sample size calculation is about 1.2 months of study time.

Bias↗
PubMedDate not supplied

Unwritten rules of talking to doctors about depression: integrating qualitative and quantitative methods.

PURPOSE: We wanted to understand concordance and discordance between physicians and patients about depression status by assessing older patient's views of interactions with their physicians. METHODS: We used an integrated mixed methods design that is both hypothesis testing and hypothesis generating. Patients aged 65 years and older, who identified themselves as being depressed, were recruited from the offices of primary care physicians and interviewed in their homes using a semistructured interview format. We compared patients whose physicians rated them as depressed with those whose physicians who did not according to personal characteristics (hypothesis testing). Themes regarding patient perceptions of their encounters with physicians were then used to generate further hypotheses. RESULTS: Patients whose physician rated them as depressed were younger than those whose physician did not. Standard measures, such as depressive symptoms and functional status, did not differentiate between patients. Four themes emerged in interviews with patients regarding how they interacted with their physicians; namely, "My doctor just picked it up," "I'm a good patient," "They just check out your heart and things," and "They'll just send you to a psychiatrist." All patients who thought the physician would "just pick up" depression and those who thought bringing up emotional content would result in a referral to a psychiatrist were rated as depressed by the physician. Few of the patients who discussed being a "good patient" were rated as depressed by the physician. CONCLUSIONS: Physicians may signal to patients, wittingly or unwittingly, how emotional problems will be addressed, influencing how patients perceive their interactions with physicians regarding emotional problems.

Age Factors↗
PubMed2000

Use of binomial group testing in tests of hypotheses for classification or quantitative covariables.

In group testing, the test unit consists of a group of individuals. If the group test is positive, then one or more individuals in the group are assumed to be positive. A group observation in binomial group testing can be, say, the test result (positive or negative) for a pool of blood samples that come from several different individuals. It has been shown that, when the proportion (p) of infected individuals is low, group testing is often preferable to individual testing for identifying infected individuals and for estimating proportions of those infected. We extend the potential applications of group testing to hypothesis-testing problems wherein one wants to test for a relationship between p and a classification or quantitative covariable. Asymptotic relative efficiencies (AREs) of tests based on group testing versus the usual individual testing are obtained. The Pitman ARE strongly favors group testing in many cases. Small-sample results from simulation studies are given and are consistent with the large-sample (asymptotic) findings. We illustrate the potential advantages of group testing in hypothesis testing using HIV-1 seroprevalence data.

Adolescent↗
PubMedDate not supplied

Structure and location of amyloid beta peptide chains and arrays in Alzheimer's disease: new findings require reevaluation of the amyloid hypothesis and of tests of the hypothesis.

New in situ high resolution electronmicroscopic examination of amyloid fibrils in situ indicate that in Alzheimer's disease these fibrils are not simply long chains of self aggregated amyloid beta peptide. The amyloid beta is not only associated with P protein and glycans, as was well known from previous immunohistologic studies, but is arranged in the form of short chains at right angles to a P protein backbone with the glycans wrapped around that backbone. These findings suggest that the hypothesis causally relating simple, fibrillar amyloid beta to Alzheimer's disease must be reevaluated since such simple fibrils may be absent, or not the major form of the amyloid beta in the brain. Other data shows that shorter multimers, so-called protofibrils, or dimers of amyloid beta or molecules cleaved from it can be highly toxic. Some of these may be in the soluble amyloid beta fraction. Shorter multimers or dimers of amyloid beta, either extra or intracellular, may be the real links between amyloid beta production and Alzheimer's disease. Toxicity studies employing fibrillar amyloid beta may not be relevant, even if they produce lesions, because they do not employ amyloid beta in the form in which it actually exists in the Alzheimer brain. Studies of treatments designed to remove fibrils or to prevent their formation may be ineffective or suboptimal in effectiveness because they do not reduce the relevant amyloid burden and/or fail to alter the arrangement of shorter multimers of amyloid beta around its P-protein and glycan core.

Alzheimer Disease↗
PubMed1996

[Homology and evolution of gene order: a simple method for testing a hypothesis on the nature of this evolution].

A method of testing various hypotheses concerning the mechanisms of evolution of gene order is suggested. Estimating the possibility of constructing an evolutionary tree that reflects the observed similarity between gene orders studied is proposed, provided that the distances between gene orders correspond to estimations obtained on the basis of the hypothesis tested. The required IBM PC software was developed. It was found that gene orders of the mouse, rabbit, cow, cat, lemur, capuchin monkey, rhesus monkey, gorilla, chimpanzee, and man could be readily interpreted in terms of the simplest ("map") model of transformation of these orders.

Animals↗
PubMed2006

Impact of criticism of null-hypothesis significance testing on statistical reporting practices in conservation biology.

Over the last decade, criticisms of null-hypothesis significance testing have grown dramatically, and several alternative practices, such as confidence intervals, information theoretic, and Bayesian methods, have been advocated. Have these calls for change had an impact on the statistical reporting practices in conservation biology? In 2000 and 2001, 92% of sampled articles in Conservation Biology and Biological Conservation reported results of null-hypothesis tests. In 2005 this figure dropped to 78%. There were corresponding increases in the use of confidence intervals, information theoretic, and Bayesian techniques. Of those articles reporting null-hypothesis testing--which still easily constitute the majority--very few report statistical power (8%) and many misinterpret statistical nonsignificance as evidence for no effect (63%). Overall, results of our survey show some improvements in statistical practice, but further efforts are clearly required to move the discipline toward improved practices.

Conservation of Natural Resources↗
PubMed2005

Testing the hypothesis of recent population expansions in nematode parasites of human-associated hosts.

It has been predicted that parasites of human-associated organisms (eg humans, domestic pets, farm animals, agricultural and silvicultural plants) are more likely to show rapid recent population expansions than are parasites of other hosts. Here, we directly test the generality of this demographic prediction for species of parasitic nematodes that currently have mitochondrial sequence data available in the literature or the public-access genetic databases. Of the 23 host/parasite combinations analysed, there are seven human-associated parasite species with expanding populations and three without, and there are three non-human-associated parasite species with expanding populations and 10 without. This statistically significant pattern confirms the prediction. However, it is likely that the situation is more complicated than the simple hypothesis test suggests, and those species that do not fit the predicted general pattern provide interesting insights into other evolutionary processes that influence the historical population genetics of host-parasite relationships. These processes include the effects of postglacial migrations, evolutionary relationships and possibly life-history characteristics. Furthermore, the analysis highlights the limitations of this form of bioinformatic data-mining, in comparison to controlled experimental hypothesis tests.

Animals↗
PubMed1998

Postoperative hepatic dysfunction in perspective. 1970.

Postoperative hepatic dysfunction will remain a difficult entity to place in perspective until increased data are obtained from prospective clinical trials. Ideally these data should compare hepatic dysfunction not only to other postoperative complications, both with regard to overall incidence and to mortality, but also to the overall risks of anesthesia and surgery. The contribution of drug-induced hepatic damage to postoperative hepatic dysfunction has remained unsettled since chloroform was first incriminated during the nineteenth century. The drug was condemned in 1912, without any attempt being made to determine the incidence of the so-called delayed chloroform poisoning, with the result that the drug is still in use and the chloroform controversy remains unsettled to this day. The halothane controversy is also unsettled and currently overshadows the former controversy, although academically of no greater importance. Although not an anesthetic, cincophen is another drug about which there is controversy concerning its hepatotoxic potential. Babior and Davidson noted that it was the first drug implicated in hepatic necrosis--presumably with the exception of chloroform--the first report appearing in 1922. In 1941 the Council on Pharmacy and Chemistry of the American Medical Association concluded that the case against cincophen was not proved and that an urgent need existed for controlled clinical studies. Twenty-five years later Babior and Davidson noted that such studies had still not been undertaken and that the situation was the same as it was a quarter of a century earlier. Perhaps the time has come for a prospective, randomized, controlled clinical trial to be undertaken so as to evaluate the hepatotoxicity of one of these drugs. Perhaps an anesthetic agent such as halothane, concerning multiple administrations of which there is currently serious question, would be a suitable choice for such a study. The drug is in wide use today, partly because of evidence of satisfactory death rates following its administration, but also because on the basis of much excellent physiological data--but an almost total lack of any confirmatory epidemiological evidence--it is thought to contribute positively toward a low overall incidence of postoperative morbidity. Perhaps in addition, as a corollary, the time has come when, as attempts to illuminate a well--enunciated problem of this nature--that is, to test a clearly formulated hypothesis--the isolated case report, the collection of isolated case reports, the series of patients reported in the absence of a proven comparable control group, and the uncontrolled survey, should be "laid to rest." At best they provide only additional hypothesis-formulating information. At worst, however, they give increased exposure to a suggestion concerning cause and effect upon which physicians may act to their patients' detriment if the hypothesis ultimately proves to be erroneous. MacMahon et al. have stated that although there is no clear-cut dividing line between descriptive and analytical epidemiology, most epidemiological studies can indeed be classified primarily as either hypothesis-formulating or hypothesis-testing. Just as we have conducted the definitive retrospective hypothesis-testing study--the National Halothane Study--demanded by the "halothane hepatitis" controversy, so must we now move to the final stage of epidemiological investigation (experimental epidemiology) by investigating the effects of multiple administrations of the drug. On this point the National Halothane Study acts more as a hypothesis-formulating study than as a hypothesis-testing study. Hill has noted that statistical problems must be dealt with by the statistical method. (ABSTRACT TRUNCATED)

Anesthetics, Inhalation↗
PubMed2003

Race, ethnicity, and depression in Canadian society.

This study examines racial/ethnic differences in mental health using data from the 1996-97 National Population Health Survey. Three hypotheses are tested. First, a socioeconomic hypothesis tests if differences in family income, education, and low income explain racial/ethnic mental health variation. Second, a social resources hypothesis tests if differences in social support explain racial/ethnic mental health variation. Finally, an interaction hypothesis tests if mental health variation stems from specific interactions of race/ethnicity with economic and social factors. Although there are socioeconomic, social resource, and interaction effects, the analysis shows that they do not fully explain racial/ethnic mental health variation. Overall, our results suggest that East and Southeast Asian, Chinese, South Asian, and black Canadians have better mental health than English Canadians. Jewish Canadians have poorer mental health than English Canadians. All other racial/ethnic groupings have similar mental health as English Canadians.

Adolescent↗
PubMed2001

On the resolution and feasibility of genome scanning approaches.

Before contemplating a genome scan to identify the map position of disease-predisposing genes, an investigator should have prior evidence of the genes' existence. It is therefore logically consistent to evaluate a genome scan experiment as an estimation problem, rather than as a hypothesis-testing problem, since absent prior evidence of the existence of disease genes, it is probably unwise to conduct the experiment at all. Recombination in a single meiosis can be modeled as a point process along the chromosome, and linkage or linkage disequilibrium (LD) mapping statistics are a simple function of the superposition of the recombination processes occurring in all meioses under study. Thus, multipoint lod scores are shown to be step functions, in the absence of ambiguity about the inheritance of chromosomal segments. The ability to map a disease gene is a function of how well the ascertained phenotypes predict the underlying trait locus genotypes. This chapter presents a thorough investigation of the properties of the multipoint lod score and uses results from renewal theory to examine the effects of deviations from a deterministic phenotype-genotype relationship. The quality of estimated gene locations is assessed through computing the mean and variance of the length of the expected 3-lod-unit support interval around the maximum likelihood estimate. The more deterministic the model, the smaller this interval is. A more exact quantification of details of this effect is used to describe the statistical properties of such genome scanning experiments from the perspective of estimation, with appropriately little regard to hypothesis testing. Hypothesis testing, however, is discussed as an appropriate context to describe linkage and LD analysis in situations where candidate genes are being screened, since only there does one have definable null and alternative hypotheses that have not been rejected before the beginning of the experiment. By contrast, it is hoped that the null hypothesis "there is no gene affecting this phenotype" has been rejected by other means before an expensive genome scan is even contemplated (though that this is often not done is probably the main problem!).

Chromosome Mapping↗
PubMedDate not supplied

[Executive functioning in unipolar depression: a review].

While several neuropsychological studies have demonstrated that cognitive deficits are seen across a broad range of cognitive domains, executive deficits associated with frontal lobe dysfunction may be prominent in depression. Executive function refers to cognitive processes that control and integrate other cognitive activities such as episodic memory. These executive functions involve a set of cognitive behaviors which include: dealing with novelty, selecting strategies, inhibiting incorrect responses, monitoring performance and using feedback to adjust future responding. The measurement of executive function relies mainly on the use of neuropsychological tests known to be sensitive to frontal lobe damage such as the Wisconsin and California Card Sorting Tests, verbal fluency tests, Stroop-test, Tower of London Task and Trail Making Test. The present review focuses on studies investigating executive functions in primary unipolar depression with these neuropsychological tasks. Unipolar depressed patients mainly exhibit cognitive inhibition deficits, problem-solving impairments and planning deficits. Cognitive inhibition deficits in depressed patients have been related to a reduction of cognitive resources and psychomotor retardation. Inhibition disturbance could lead depressed patients to process irrelevant information and consequently reduce their capacity to control transient mood changes. Several studies have found evidence of problem solving impairments in depressed patients. Depressed subjects show with card sorting tests difficulties in hypothesis testing with a loss of spontaneous and reactive cognitive flexibility. The cognitive rigidity and hypothesis-testing associated with dorsolateral prefrontal dysfunction in depression may prevent patients to cope with life events and lead to a perpetuation of depressed mood by a continuation of stress exposure. Planning tasks, such as the Tower of London Test, also demonstrate that depressed patients fail to use negative feedback as a motivational boost to improve their performance. Both trait and state factors influence the executive level of depressed patients. Executive deficits have been reported in more severely depressed subjects with melancholic or psychotic features. Executive functioning also might predict a poorer outcome in depression. Thus initiation and perseveration scores - a measure of cognitive flexibility - is associated with relapse and recurrence of depression and residual depressive symptoms. Brain imaging studies show that reduced blood flow, particularly in medial prefrontal cortex and dorsal anterior cingulate cortex subserve executive impairments in depression. However neuroimaging studies underscore the importance of mood-cognitive interactions in depression. A recent working model of depression (Mayberg et al., 1999) implicates failure of the coordinated interactions of distributed cortical-limbic pathways in the neuropsychopathology of depression. According to this model, neocortical (prefrontal and parietal regions) and superior limbic elements (dorsal anterior cingulate) are postulated to mediate impaired attention and executive function, whereas ventral limbic regions (ventral anterior cingulate, subcortical structures) are postulated to mediate circadian and vegetative aspects of depression. Further studies are needed to validate this model at the neuropsychological level as well as the brain level and to elucidate the complex interactions between mood, cognitive resources and executive function in depression.

Cognition Disorders↗
PubMed2004

In support of null hypothesis significance testing.

Many criticisms have been levelled at null hypothesis significance testing (NHST). It is argued here that although there is reason to doubt that data subjected only to NHST have been subjected to sufficient analysis, the search for clear answers to well-formulated questions derived from substantive hypotheses is well served by NHST. To reliably draw inferences from data, however, NHST may need to be complemented by additional methods of analysis, such as the use of confidence intervals and of estimates of the degree of association between independent and dependent variables. It is argued that these should be seen as complements of, rather than as substitutes for, NHST since they do not directly test the strength of evidence against a null hypothesis.

Confidence Intervals↗
PubMed2000

Effects of topiramate on kainate- and domoate-activated [14C]guanidinium ion flux through GluR6 channels in transfected BHK cells using Cytostar-T scintillating microplates.

PURPOSE: This study was undertaken to test the hypothesis that topiramate (TPM) exerts a negative modulatory effect on some types of alpha-amino-3-hydroxy-5-methylisoxazole-4-proprionic acid (AMPA)/kainate receptors by binding to the site at which protein kinase A (PKA) phosphorylates the receptor-channel complex. METHODS: The effect of TPM on kainate- or domoate-induced [14C]guanidinium ion flux through iGluR6 channels expressed in baby hamster kidney (BHK) cells was evaluated. Because the hypothesis predicts that TPM will bind only in the dephosphorylated state, a variety of experimental conditions were used to either promote or impede the phosphorylation of the receptor-channel complex. These included the use of dibutyryl cyclic adenosine monophosphate (cAMP) and forskolin to activate PKA, H-9 and H-89 to inhibit PKA, and okadaic acid to inhibit protein phosphatases. RESULTS: Kainate (1 microM) induced a gradual accumulation of [14C]guanidinium into the cells that plateaued approximately 30 min after initiation of the reaction, whereas domoate (0.1 microM) caused a rapid accumulation into the cells that peaked within 5 min; thereafter, the amount of [14C]guanidinium in the cells declined gradually. Topiramate, at 0.1 and 100 microM, did not significantly affect the [14C]guanidinium accumulation under any of the experimental conditions used. CONCLUSIONS: The results of this study are not consistent with the hypothesis tested. However, the results must be interpreted cautiously because iGluR6 receptors expressed in the BHK cells and the functional state of proteins that regulate AMPA/receptors (e.g., PSD-95) may not be sufficiently similar to the receptors and functional state in neurons to serve as a true test of the hypothesis.

Animals↗
PubMed2000

Null hypothesis significance testing: a review of an old and continuing controversy.

Null hypothesis significance testing (NHST) is arguably the most widely used approach to hypothesis evaluation among behavioral and social scientists. It is also very controversial. A major concern expressed by critics is that such testing is misunderstood by many of those who use it. Several other objections to its use have also been raised. In this article the author reviews and comments on the claimed misunderstandings as well as on other criticisms of the approach, and he notes arguments that have been advanced in support of NHST. Alternatives and supplements to NHST are considered, as are several related recommendations regarding the interpretation of experimental data. The concluding opinion is that NHST is easily misunderstood and misused but that when applied with good judgment it can be an effective aid to the interpretation of experimental data.

Data Interpretation, Statistical↗
PubMed1981

Multivariate logit analysis of concordance ratios for quantitative traits in twin studies.

The application of multiway contingency table analysis to the multivariate analysis of concordance ratios in twin studies is developed. The approach is illustrated by data on smoking and alcohol use in Finland and Sweden. This approach can enable the assessment of the effect of other variables on the concordance ratio and thus allow estimates of genetic effects on the trait under study. Hypotheses on relationships between genetic effects and other variables can be tested. After hypothesis testing, model fitting of the best hypothesis can be carried out.

Adult↗
PubMed2003

The design and analysis of microarray experiments: applications in parasitology.

Microarray experiments can generate enormous amounts of data, but large datasets are usually inherently complex, and the relevant information they contain can be difficult to extract. For the practicing biologist, we provide an overview of what we believe to be the most important issues that need to be addressed when dealing with microarray data. In a microarray experiment we are simply trying to identify which genes are the most "interesting" in terms of our experimental question, and these will usually be those that are either overexpressed or underexpressed (upregulated or downregulated) under the experimental conditions. Analysis of the data to find these genes involves first preprocessing of the raw data for quality control, including filtering of the data (e.g., detection of outlying values) followed by standardization of the data (i.e., making the data uniformly comparable throughout the dataset). This is followed by the formal quantitative analysis of the data, which will involve either statistical hypothesis testing or multivariate pattern recognition. Statistical hypothesis testing is the usual approach to "class comparison," where several experimental groups are being directly compared. The best approach to this problem is to use analysis of variance, although issues related to multiple hypothesis testing and probability estimation still need to be evaluated. Pattern recognition can involve "class prediction," for which a range of supervised multivariate techniques are available, or "class discovery," for which an even broader range of unsupervised multivariate techniques have been developed. Each technique has its own limitations, which need to be kept in mind when making a choice from among them. To put these ideas in context, we provide a detailed examination of two specific examples of the analysis of microarray data, both from parasitology, covering many of the most important points raised.

Analysis of Variance↗
PubMed2004

Do males have a better chance of mating when the number of estrous females is equal to or greater than the males' ordinal rank? Testing the hypothesis in Japanese macaques.

This study was designed to test the hypothesis that male primates in multi-male/multi-female social groups with a clear male dominance hierarchy have a better chance of mating when the number of estrous females is equal to or greater than, as opposed to less than, the males' ordinal rank. I studied a Japanese macaque (Macaca fuscata fuscata) troop during mating seasons from 1992 to 1995. The mean daily operational sex ratio (OSR; the number of estrous females per troop male), which was calculated on observation days, was 0.21, 1.9, 0.48, and 3.1 in 1992-1995, respectively. Overall, focal animal sampling of males yielded 118 male-day records. The male-day records for each male were divided into the two estrous female number conditions: 1) the male-day records when the number of estrous females was equal to or greater than the male's ordinal rank, and 2) the male-day records when the number of estrous females was less than the male's ordinal rank. In the 1993 and 1995 mating seasons, when the number of estrous females was equal to or greater than the ordinal rank of each male, all of the males were observed mating. Conversely, when the number of estrous females was less than the ordinal rank of some male, they were not observed mating in the 1992 and 1994 mating seasons. The percentage for each male across each male's total mating opportunity was <20% when the number of estrous females was less than the male's ordinal rank. By contrast, the percentage for each male across each male's total mating opportunity exceeded 45% when the number of estrous females was equal to or greater than the male's ordinal rank, except for one male. Of all the male-day records for males observed mating with ejaculation, 41 were obtained when the number of estrous females was equal to or greater than the male's rank; conversely, only three records were obtained when the number of estrous females was less than the male's ordinal rank. Therefore, it appears that males have a better chance of mating when the number of estrous females is equal to or greater than the males' ordinal rank, as opposed to when the number is less than their ordinal rank.

Animals↗