Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Statistical power analysis”

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
PubMed2000

[Evidence-based medicine. Input of epidemiologic studies].

Epidemiological studies, more specifically those related to analytical epidemiology are a major determinant of evidence-based medicine. When classifying the value of the different epidemiological studies, based on their level of causality, cohort studies are ranked higher than case-control studies, mainly due to the fact that they allow direct collection of information on exposition to risk factors and health consequences. However, case-control studies can also provide an important information, if a specific effort is dedicated to the analyses of the circumstances of the exposition. Meta-analysis increases, by pooling them, the statistical power of individual studies of limited size. Meta-analysis can also be considered as an important source of evidence in the perspective of evidence-based medicine.

Case-Control Studies↗
PubMed2006

Rosetta error model for gene expression analysis.

MOTIVATION: In microarray gene expression studies, the number of replicated microarrays is usually small because of cost and sample availability, resulting in unreliable variance estimation and thus unreliable statistical hypothesis tests. The unreliable variance estimation is further complicated by the fact that the technology-specific variance is intrinsically intensity-dependent. RESULTS: The Rosetta error model captures the variance-intensity relationship for various types of microarray technologies, such as single-color arrays and two-color arrays. This error model conservatively estimates intensity error and uses this value to stabilize the variance estimation. We present two commonly used error models: the intensity error-model for single-color microarrays and the ratio error model for two-color microarrays or ratios built from two single-color arrays. We present examples to demonstrate the strength of our error models in improving statistical power of microarray data analysis, particularly, in increasing expression detection sensitivity and specificity when the number of replicates is limited.

Algorithms↗
PubMed1991

Meta-analysis.

Meta-analysis corresponds to all systematic methods which use statistical techniques for combining results from several independent studies. The aim is to get a consistent estimation of the global effect of a procedure on a specified outcome. The technique allows us to increase the power of statistical testing, and to get information which cannot be drawn from one individual study. Two approaches are possible, and often combined: the qualitative approach consists of weighing various studies according to their methodological quality; the quantitative approach consists of pooling the results of different studies, in order to generate results with a higher statistical power. A meta-analysis is a long and rigorous process, which follows several steps: statement of objectives; definition of articles inclusion and exclusion criteria; literature search; collection of data and evaluation of the quality of each study; tests for homogeneity; pooling; sensitivity analyses; presentation of results; and conclusions. Meta-analysis has several advantages: it estimates the size of an effect; it improves the generalizability; it compels to rigor; it lessens the part of subjectivity. When meta-analysis is cautiously and properly done, it brings new useful information, and helps physicians and health policy makers in answering to a specific question.

Data Collection↗
PubMed2000

Measuring morbidity: disease counts, binary variables, and statistical power.

OBJECTIVES: This study compares the use of the binary disease variables with counts of the same conditions in models of self-rated health to better understand the advantages and disadvantages of each approach. In particular, the analysis seeks to determine if statistical power is adequate for the binary variable approach. METHODS: Morbidity measures from adults in 2 large national surveys were used in both cross-sectional and longitudinal analyses. RESULTS: Although differences across the approaches are modest, the binary variable approach offers greater explanatory power and slightly higher R2 values. Despite these advantages, statistical power is insufficient in some cases, especially for conditions that are relatively rare and/or that manifest modest differences on the outcome variable. DISCUSSION: Statistical power estimates are advisable when using the binary variable approach, especially if the list of diseases and health conditions is extensive. Although a simple count of diseases may be useful in some research applications, separate counts for serious and nonserious conditions should be more useful in many research projects while avoiding the risk of inadequate statistical power.

Adult↗
PubMed1993

Bioavailability of temazepam: comparison of four 7.5-MG capsules with a single 30-MG capsule.

OBJECTIVE: To compare the bioavailability of four temazepam 7.5-mg capsules (Restoril, Sandoz Pharmaceuticals) with that of a single temazepam 30-mg capsule. DESIGN: Single-dose, open-label, two-period, crossover (replicated Latin square). SETTING: Domiciled environment for clinical testing. PARTICIPANTS: Twenty-six healthy male volunteers aged 18-40 years; 25 completed the study. INTERVENTIONS: Subjects randomly received either four temazepam 7.5-mg capsules or one temazepam 30-mg capsule. Blood samples were drawn at various time points after each period (0-48 h), and analyzed for plasma concentration of temazepam. The washout period between doses was five days. MAIN OUTCOME MEASUREMENTS: Five parameters of both dosage forms were compared: (1) area under curve (AUC), (2) peak concentration (Cmax), (3) time to peak concentration (Tmax), (4) apparent rate constant for absorption, and (5) lag time for appearance of drug in plasma. Statistical procedures included ANOVA, power analysis, and confidence limits. RESULTS: The mean AUC for the four 7.5-mg capsules and one 30-mg capsule differed by less than 2 percent and the mean Cmax differed by less than 14 percent for the two dosage strengths; neither of these differences reached statistical significance (p > 0.05). The 7.5-mg capsules reached peak plasma concentrations significantly faster than the 30-mg dosage form (mean Tmax 1.18 and 1.73 h, respectively; p = 0.01). CONCLUSIONS: The two formulations of temazepam were bioequivalent with respect to the extent of bioavailability. Regarding the rate of absorption, however, the 7.5-mg capsules reached peak plasma concentrations significantly faster than the 30-mg dosage form.

Adolescent↗
PubMed1999

Tools of the trade: statistical analysis in dermatopathology articles.

Statistical analysis of research results provides powerful tools for understanding the data from projects. A prospective review of the statistical methods utilized in 100 consecutive articles in the recent literature relevant to dermatopathology was performed. The majority of papers, 75%, did not contain statistical analyses. A wide variety of methods were utilized in the papers that did have statistical analyses including methods in the following categories: t-test, contingency tables, multiple regression, multiple comparisons, nonparametric tests, life tables, and survival tests. Nine of the 25 papers utilizing statistical analysis had problems in the methods. These problems included treating categorical data as a continuous variable, multiple comparisons, subgroup analysis, and discordance of statistics and conclusions. It is important to understand the underlying assumptions of statistical methods to use the appropriate tets. These are tools of the trade for dermatopathology investigators.

Data Interpretation, Statistical↗
PubMed2004

Otitis media and speech and language: a meta-analysis of prospective studies.

OBJECTIVE: Considerable controversy surrounds whether a history of otitis media with effusion (OME) in early childhood causes later speech and language problems. We conducted a meta-analysis of prospective studies to determine: 1) whether a history of OME in early childhood is related to receptive language, expressive language, vocabulary, syntax, or speech development in children 1 to 5 years old and 2) whether hearing loss caused by otitis media in early childhood is related to children's receptive language or expressive language through 2 years of age. METHODS: We searched online databases and bibliographies of OME studies and reviews for prospective or randomized clinical trials published between January 1966 and October 2002 that examined the relationship of OME or OME-associated hearing loss in early childhood to children's later speech and language development. The original search identified 38 studies, of which 14 had data suitable for calculating a pooled correlation coefficient (correlational studies) or standard difference between parallel groups (group studies). Random-effects meta-analysis was used to pool data when at least 3 studies had usable data for a particular outcome. RESULTS: We performed 11 meta-analyses. There were no significant findings for the analyses of OME during early childhood versus receptive or expressive language during the preschool years in the correlation studies. Similarly, there were no significant findings for OME versus vocabulary, syntax, or speech during the preschool years. Conversely, there was a significant negative association between OME and preschoolers' receptive and expressive language (lower language) (0.24 and 0.25 standard difference, respectively) in the group studies. Additionally, hearing was also related to receptive and expressive language in infancy (3%-9% of variance). CONCLUSIONS: Our results indicate no to very small negative associations of OME and associated hearing loss to children's later speech and language development. These findings may overestimate the impact of OME on outcomes, because most studies did not adjust for known confounding variables (such as socioeconomic status) and excluded data not suitable for statistical pooling, especially from methodologically sound studies. Although some OME language differences were detectable by meta-analysis due to increased statistical power, the clinical relevance for otherwise healthy children is uncertain.

Child, Preschool↗
PubMed2004

Flexural strength and modulus of a step-cured resin composite.

In an attempt to reduce marginal contraction gaps, a step-curing mode of light polymerization of resin composite has been proposed. It was hypothesized that such an exposure mode, having an initial reduced curing rate, would result in composite having a lower modulus of elasticity than composite exposed in one step. The composite was initially exposed to power density levels of 50, 100, 150, or 200 mW/cm2 for durations of 10, 20, or 40 s. The final exposure and the exposure of the control group were performed at 750 mW/cm2 for 20 s. It was found that certain modes of two-step exposure resulted in a lower flexural modulus than did a one-step exposure of constant, high power density. Regression analysis showed, with statistical significance, that flexural modulus was relatively high following a short initial exposure at low power density or using a long initial exposure at high power density. Conversely, flexural modulus was relatively low following a long initial exposure at low power density or using a short initial exposure at high power density. The results were explained by probable differences in degree of conversion and cross-link density of the polymer network.

Analysis of Variance↗
PubMed2004

Is categorization performance really intact in amnesia? A meta-analysis.

Most published studies of category learning in amnesia have reported intact categorization performance. These results have been used to challenge single-system accounts of categorization and recognition, in which a single representational system mediates performance in these two tasks. Many of the published studies, however, have shown a numerical advantage for controls over amnesics and often have had low statistical power. A meta-analysis was conducted to assess whether this numerical advantage is significant when the data are pooled across studies. This analysis indicates that amnesic subjects do, in fact, show deficits in categorization tasks, which is consistent with single-system exemplar model predictions.

Amnesia↗
PubMed1989

Translating a health professional certification test to another language. A pilot analysis.

There has been increased pressure in recent years for examinations used for occupational and professional credentialing to be offered in several languages. When credentialing programs are international in scope, rather than national, examination translation may be a necessity, and a validation of the translation will be required. Yet, few candidates may sit for the translated version, jeopardizing the power and generalizability of statistical analysis. How should a board proceed in meeting its linguistic responsibility when faced with a small n? This article describes such a situation, and the linguistic and psychometric procedures that were developed and applied on a pilot translation by the International Board of Lactation Consultant Examiners (IBLCE).

Allied Health Personnel↗
PubMed2000

Comparison of an extract of hypericum (LI 160) and sertraline in the treatment of depression: a double-blind, randomized pilot study.

BACKGROUND: Hypericum (St. John's wort) has been shown to be as efficacious and well tolerated as standard antidepressants in the treatment of depression but has not been compared with selective serotonin reuptake inhibitors (SSRIs). OBJECTIVE: This study compared hypericum and the SSRI sertraline in the treatment of depression. METHODS: In a double-blind, randomized study conducted in a community hospital, 30 male and female outpatients (19 women, 11 men; mean age, 45.5 years) with mild to moderate depression received 600 mg/d of a standardized extract of hypericum (LI 160) or 50 mg/d sertraline for I week, followed by hypericum 900 mg/d or sertraline 75 mg/d for 6 weeks. RESULTS: The severity of symptoms, as assessed by scores on the Hamilton Rating Scale for Depression (HAM-D) and the Clinical Global Impression scale, was significantly reduced in both treatment groups (P < 0.01). Clinical response (defined as a > or =50% reduction in HAM-D scores) was noted in 47% of patients receiving hypericum and 40% of those receiving sertraline. The difference was not statistically significant. Both agents were well tolerated. A post hoc power analysis indicated that failure to reach statistical significance between treatments resulted primarily from an absence of clinical differences rather than the small sample size. CONCLUSION: The hypericum extract was at least as effective as sertraline in the treatment of mild to moderate depression in a small group of outpatients.

Adolescent↗
PubMed1999

Elevated risk of cancer of the urinary tract for alcohol drinkers: a meta-analysis.

OBJECTIVES: Recent narrative reviews have concluded that there is no support for an association between alcohol consumption and urinary tract cancer. Many individual studies, however, have reported positive associations, although rarely statistically significant. The purpose of this meta-analysis is to summarize and quantify this relationship with more statistical power and to perform a sensitivity analysis on the study characteristics. METHODS: We included 16 epidemiological studies published up to April 1999 and calculated summary odds ratios (SORs), both upgraded and adjusted for age, sex and smoking by meta-regression analyses. The age- and smoking-adjusted SORs (current alcohol drinking vs. non-drinking) were 1.3 (95% CI 0.9-2.0) for six studies with men and 1.0 (95% CI 0.4-2.6) for four studies with women. RESULTS: The age-, sex- and smoking-adjusted SOR was 1.2 (95% CI 0.9-1.7) for seven studies with men and women combined. CONCLUSION: Even though studies differed in methodology, the results were rather consistent. Subgroup analyses by type or amount of alcohol were not possible due to sparse data. We conclude that the available data suggest a slightly increased risk of urinary tract cancer from alcohol consumption for men. The risk related to alcohol consumption for women and the influence of the amount and type of alcohol remain unclear.

Adult↗
PubMed2006

An interactive power analysis tool for microarray hypothesis testing and generation.

MOTIVATION: Human clinical projects typically require a priori statistical power analyses. Towards this end, we sought to build a flexible and interactive power analysis tool for microarray studies integrated into our public domain HCE 3.5 software package. We then sought to determine if probe set algorithms or organism type strongly influenced power analysis results. RESULTS: The HCE 3.5 power analysis tool was designed to import any pre-existing Affymetrix microarray project, and interactively test the effects of user-defined definitions of alpha (significance), beta (1-power), sample size and effect size. The tool generates a filter for all probe sets or more focused ontology-based subsets, with or without noise filters that can be used to limit analyses of a future project to appropriately powered probe sets. We studied projects from three organisms (Arabidopsis, rat, human), and three probe set algorithms (MAS5.0, RMA, dChip PM/MM). We found large differences in power results based on probe set algorithm selection and noise filters. RMA provided high sensitivity for low numbers of arrays, but this came at a cost of high false positive results (24% false positive in the human project studied). Our data suggest that a priori power calculations are important for both experimental design in hypothesis testing and hypothesis generation, as well as for the selection of optimized data analysis parameters. AVAILABILITY: The Hierarchical Clustering Explorer 3.5 with the interactive power analysis functions is available at www.cs.umd.edu/hcil/hce or www.cnmcresearch.org/bioinformatics. CONTACT: jseo@cnmcresearch.org

Algorithms↗
PubMed1999

Association between prolactin and disease activity in systemic lupus erythematosus. Influence of statistical power.

OBJECTIVE: To analyze the statistical power of studies in the medical literature on the relationship between prolactin (PRL) and systemic lupus erythematosus (SLE) activity. METHODS: Published studies that were identified through MEDLINE search, as well as references from these articles, were reviewed. RESULTS: We identified 5 articles that sought to establish a relationship between PRL and SLE activity. In 4 of them, the frequency of hyperprolactinemia in SLE (>20 ng/ml) was 2.2-47.2%, and in one article, there was a relationship between PRL and SLE activity. A power analysis of individual studies could be carried out in only 2 of the 5; there were no significant effects; the 2 articles cited differences in the frequency of hyperprolactinemia in patients with and without lupus activity (1.6 and 12.3%, respectively), but because of a low power of the studies (> or =30.8%), it could not be determined whether the differences in the frequency of hyperprolactinemia were significant. On the other hand, joint analysis of 3 articles showed a significant association between hyperprolactinemia and lupus activity. CONCLUSION: Published clinical results concerning the relationship between PRL and Jupus activity are contradictory, due in part to the statistical power of the studies. Our analysis of these studies showed that PRL is related to lupus activity, without establishing a formal causal relationship.

Adolescent↗
PubMed2005

Exploring whole genome amplification as a DNA recovery tool for molecular genetic studies.

The whole genome amplification (WGA) protocol evaluated during this study, GenomiPhi DNA amplification kit, is a novel method that is not based on polymerase chain reaction but rather relies on the highly processive and high fidelity Phi29 DNA polymerase to replicate linear genomic DNA by multiple strand displacement amplification. As little as 1 ng of genomic DNA template is sufficient to produce microgram quantities of high molecular weight DNA. The question explored during this study is whether such a WGA method is appropriate to reliably replenish and even recover depleted DNA samples that can be used for downstream genetic analysis. A series of human DNA samples was tested in our laboratory and validated using such analytical methods as gene-specific polymerase chain reaction, direct sequencing, microsatellite marker analysis, and single nucleotide polymorphism allelic discrimination using TaqMan and Pyrosequencing chemistries. Although degraded genomic DNA is not a good template for Phi29 WGA, this method is a powerful tool to replenish depleted DNA stocks and to increase the amount of sample for which biological tissue availability is scarce. The testing performed during the validation phase of the study indicates no discernable difference between WGA samples and the original DNA templates. Thus, GenomiPhi WGA can be used to increase precious or depleted DNA stocks, thereby extending the life of a family-based linkage analysis project and increasing statistical power.

Adult↗
PubMed2002

Stochastic processes defining sensitivity and variability of internally calibrated quantitative NASBA-based viral load assays.

For quantitative assessment of virus particles in patient plasma samples various assays are commercially available. Typical performance characteristics for such assays are sensitivity, precision and the range of linearity. In order to assess these properties it is common practice to divide the range of inputs into subranges in order to apply different statistical models to evaluate these properties separately. We developed a general statistical model for internally calibrated amplification based viral load assays that combines these statistical properties in one powerful analysis. Based on the model an unambiguous definition of the lower limit of the linear range can be given. The proposed method of analysis was illustrated by a successful application to data generated by the NucliSens EasyQ HIV-1 assay.

HIV-1↗