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 847 records · Page 47Linked to original sources
PubMed2000

The application of multilevel, multivariate modelling to orthodontic research data.

OBJECTIVE: To demonstrate the use of multilevel multivariate modelling in the evaluation of multiple outcome dental data. BASIC RESEARCH DESIGN: Multiple outcome dental research data are used to illustrate the problems of analysing such complex information structures i.e. several outcomes clustered within subjects. Appropriate and statistically efficient methods of data analysis are proposed and illustrated step-by-step. The data structure is analysed using multilevel multivariate regression techniques and this process is discussed in comparison to conventional single-level multiple regression. PARTICIPANTS: Questionnaire data were obtained from an orthognathic study of 84 subjects seeking treatment and 106 'non-treatment' controls (full details of which are reported elsewhere). RESULTS: Multivariate multiple regression analysis demonstrated a number of advantages over separate single-level multiple regression approaches, including a gain in statistical efficiency and greater insight into: a) the role of (significant) explanatory variables and b) outcome variable interactions. Multilevel multivariate analysis reduced the risk of both Tipe I and Type II statistical errors. CONCLUSIONS: The study demonstrates the benefit of multilevel multivariate modelling over conventional single-level techniques for statistical analysis of multiple outcome data. As a result of ongoing technical developments in the power, speed and memory of modern PCs, multilevel multivariate regression can now be undertaken with relative ease. Consequently, researchers are better equipped to analyse such complex data structures, particularly within dentistry where multivariate data are common.

Data Interpretation, Statistical↗
PubMed2006

"Hang Up Your Pocketbook" -- an easy intervention for the granny syndrome: grandparents as a risk factor in unintentional pediatric exposures to pharmaceuticals.

CONTEXT: Although the circumstances are not well studied, grandparents' medications account for 10% to 20% of unintentional pediatric intoxications in the United States. OBJECTIVES: To characterize circumstances leading to and outcomes from pediatric pharmaceutical exposures. To identify preventable risk factors associated with this pattern of injury, referred to as the "granny syndrome." DESIGN, SETTING, AND PARTICIPANTS: Retrospective review of records of telephone calls made to a certified regional poison control center in the United States. Records were analyzed for all calls concerning children aged 6 years or younger who were exposed to grandparents' medication(s). For statistical analysis, regression and chi(2) analysis as well as Fisher exact tests were used. The sample size provided 80% power to detect a 10% difference at the 5% level of significance. Statistical significance was set at P< or =.05. MAIN OUTCOMES MEASURED: Use of child-resistant containers (CRCs), the location of pharmaceuticals prior to pediatric ingestion, and drug classes involved (eg, analgesics, cardiovascular drugs). RESULTS: Of the 200 incidents analyzed, 90 (45%) cases involved CRCs, and 110 (55%) involved containers that were not child resistant. For these incidents, the average age of the child was 18.8 months; the grandparent was aged on average 58.7 years. Most medications had been placed on tables or countertops (91 [46%]), low shelves (57 [29%]), or in pocketbooks (34 [17%]). The type of container in which the pharmacologic agent was stored (CRC vs non-CRC) was not statistically significant (P>.1). Ease of access to medication, regardless of the type of container used, was the only statistically significant outcome (P<.001). In the present study, accidental pediatric exposures most frequently involved cardiovascular (90 [45%]), analgesic (84 [42%]), and psychotropic (32 [16%]) medications. CONCLUSION: Pediatric exposure to pharmaceutical agents is a preventable cause of injury. Physicians have an important opportunity to assist in preventing pediatric pharmaceutical exposures by instructing parents and grandparents on how to better limit children's access to medications as an essential component to enhance child safety.

Accidents, Home↗
PubMed2007

Do polymorphisms of 5,10-methylenetetrahydrofolate reductase (MTHFR) gene affect the risk of childhood acute lymphoblastic leukemia?

Meta-analysis has become an important statistical tool in genetic association studies, since it may provide more powerful and precise estimates. However, meta-analytic studies are prone to several potential biases not only because the preferential publication of "positive'' studies but also due to difficulties in obtaining all relevant information during the study selection process. In this letter, we point out major problems in meta-analysis that may lead to biased conclusions, illustrating an empirical example of two recent meta-analyses on the relation between MTHFR polymorphisms and risk of acute lymphoblastic leukemia that, despite the similarity in statistical methods and period of study selection, provided partially conflicting results.

Bias↗
PubMed2006

Small-scale randomized controlled trials need more powerful methods of mediational analysis than the Baron-Kenny method.

OBJECTIVE: To devise more-effective physical activity interventions, the mediating mechanisms yielding behavioral change need to be identified. The Baron-Kenny method is most commonly used, but has low statistical power and may not identify mechanisms of behavioral change in small-to-medium size studies. More powerful statistical tests are available. STUDY DESIGN AND SETTING: Inactive adults (N=52) were randomized to either a print or a print-plus-telephone intervention. Walking and exercise-related social support were assessed at baseline, after the intervention, and 4 weeks later. The Baron-Kenny and three alternative methods of mediational analysis (Freedman-Schatzkin; MacKinnon et al.; bootstrap method) were used to examine the effects of social support on initial behavior change and maintenance. RESULTS: A significant mediational effect of social support on initial behavior change was indicated by the MacKinnon et al., bootstrap, and, marginally, Freedman-Schatzkin methods, but not by the Baron-Kenny method. No significant mediational effect of social support on maintenance of walking was found. CONCLUSIONS: Methodologically rigorous intervention studies to identify mediators of change in physical activity are costly and labor intensive, and may not be feasible with large samples. The use of statistically powerful tests of mediational effects in small-scale studies can inform the development of more effective interventions.

Aged↗
PubMed2004

Epidemiology of breastfeeding: advances and multidisciplinary applications.

Advancing understanding of the importance of breastfeeding (BF) has required advances in application of epidemiologic methods to BF research. This chapter considers, in the context of BF epidemiology, how different research designs complement each other to establish firmer causal inferences, the need for multidisciplinary collaborations, and the application of powerful analytical methods to strengthen epidemiologic studies. A recent global meta-analysis of 47 studies documenting the inverse association between lifetime BF duration and breast cancer risk illustrates the value of longitudinal (cohort) studies to establish the temporal sequence of events and to rule out differential recall biases that can threaten the internal validity of case-control studies. The discovery of the transmission of human immunodeficiency virus (HIV) via human milk has led epidemiologists to provide important hypotheses to be tested at the molecular and cellular level and in turn test hypotheses generated in laboratory studies. Application of powerful statistical methods to BF epidemiology has helped strengthen causal inferences. A retrospective analysis of the 1976--1977 Malaysian Family Life Survey found interactive multivariate statistical models to better understand how different socioeconomic and cultural contexts modify the influence of BF on child health. Further, a study conducted in Cebu City, Philippines, demonstrated how structural equation models can be used to test the pathways through which proximate (e.g., BF) and distal (e.g., socioeconomic status) determinants affect child health after adjusting for reverse causality (i.e., child health influencing maternal BF choice). For ethical reasons, BF randomized controlled trials (RCTs) have seldom been conducted. Recent RCTs examining the link between BF peer counseling and BF outcomes in diverse settings provides a unique opportunity to firmly establish the causal inferences regarding BF and the prevention of diarrheal diseases. The value of rapid assessment anthropologic methods to guide development of peer counseling interventions subsequently tested through RCTs is presented. This chapter also illustrates how Cox regression and logistic regression can be applied to BF research.

Breast Feeding↗
PubMed2000

An evaluation of homogeneity tests in meta-analyses in pain using simulations of individual patient data.

In this paper we consider the validity and power of some commonly used statistics for assessing the degree of homogeneity between trials in a meta-analysis. We show, using simulated individual patient data typical of that occurring in randomized controlled trials in pain, that the most commonly used statistics do not give the expected levels of statistical significance (i.e. the proportion of trials giving a significant result is not equal to the proportion expected due to random chance) when used with truly homogeneous data. In addition, all such statistics are shown to have extremely low power to detect true heterogeneity even when that heterogeneity is very large. Since, in most practical situations, failure to detect heterogeneity does not allow us to say with any helpful degree of certainty that the data is truly homogeneous, we advocate the quantitative combination of results only where the trials contained in a meta-analysis can be shown to be clinically homogeneous. We propose as a definition of clinical homogeneity that all trials have (i) fixed and clearly defined inclusion criteria and (ii) fixed and clearly defined outcomes or outcome measures. In pain relief, for example, the first of these would be satisfied by all patients having moderate or severe pain, whilst the second would be satisfied by using at least 50% pain relief as the successful outcome measure.

Algorithms↗
PubMed2005

Efficiency and power in genetic association studies.

We investigated selection and analysis of tag SNPs for genome-wide association studies by specifically examining the relationship between investment in genotyping and statistical power. Do pairwise or multimarker methods maximize efficiency and power? To what extent is power compromised when tags are selected from an incomplete resource such as HapMap? We addressed these questions using genotype data from the HapMap ENCODE project, association studies simulated under a realistic disease model, and empirical correction for multiple hypothesis testing. We demonstrate a haplotype-based tagging method that uniformly outperforms single-marker tests and methods for prioritization that markedly increase tagging efficiency. Examining all observed haplotypes for association, rather than just those that are proxies for known SNPs, increases power to detect rare causal alleles, at the cost of reduced power to detect common causal alleles. Power is robust to the completeness of the reference panel from which tags are selected. These findings have implications for prioritizing tag SNPs and interpreting association studies.

Algorithms↗
PubMed1984

A three component alcohol attitude scale.

A summated rating scale of the Likert-type was constructed to measure college students' attitudes toward alcohol use and abuse. Using a three-component approach, a table of specifications for the instrument was developed to include three subscales to measure feeling, belief and intention to act. Comparable emphasis in each subscale was given to drinking behavior, results of drinking and social perspectives. After extensive item development, two experimental forms were administered to a sample of undergraduate college students to gather evidence of item function and validity. Statistical analysis produced a pool of items meeting criteria for internal consistency and discrimination power. According to the table of specifications, 54 items were selected to provide 18 conceptually comparable items for each subscale in the final instrument. The scale was then administered to a sample of 880 undergraduate students at three large, midwestern universities. The data were analyzed statistically to test nine hypotheses concerning validity and performance of the scale. Results demonstrated highly significant levels of internal consistency and discriminating power of items, the subscales and the total scale. The three-component approach was found to be an appropriate, effective technique for measuring attitudes toward health-related behavior.

Adult↗
PubMed2026

Phylogenomic subsampling and upsampling for efficient evolutionary analyses of big data.

Long runtimes, high memory demands, and reliance on high-performance computing impede phylogenomic analyses. We review a scalable phylogenomic subsampling with upsampling (PSU) framework to address this challenge, which reduces runtime and memory requirements by orders of magnitude. In PSU, small subsamples of sites from a concatenated alignment are analyzed, which are expanded by upsampling before inference, and the resulting inferences are aggregated to obtain evolutionary estimates. PSU harnesses the fact that the computational cost of maximum likelihood analysis is strongly influenced by the number of distinct site patterns in the concatenated alignment, whereas statistical power depends primarily on the amount of evolutionary information represented by the total number of sites and substitutions. By reducing the former while restoring the latter through upsampling, PSU can approximate many full-alignment analyses at substantially lower computational cost. Analysis of simulated and empirical datasets shows that PSU can accurately estimate bootstrap support values, select the optimal substitution model, test evolutionary hypotheses, and infer branch lengths, divergence times, and associated uncertainty measures. PSU also provides distributions of inferred clade support across independent subsamples, enabling detection of conflicting phylogenetic signals that may remain hidden in conventional bootstrap analysis of concatenated alignments. Automated tuning of subsample size, the number of subsamples, and the number of upsampling replicates make PSU practical. We suggest that PSU is a general approach for scalable phylogenomic inference using a broad range of statistical methods. By enabling analyses of genome-scale alignments on commodity hardware, PSU broadens research access and reduces environmental and infrastructural costs of big-data phylogenomics.

Phylogeny↗
PubMed2026

Phylogenomic subsampling and upsampling for efficient evolutionary analyses of big data.

Long runtimes, high memory demands, and reliance on high-performance computing impede phylogenomic analyses. We review a scalable phylogenomic subsampling with upsampling (PSU) framework, in which small subsamples of sites from a concatenated alignment are expanded by upsampling before inference, and the resulting analyses are then aggregated to obtain evolutionary estimates. PSU harnesses the fact that the computational cost of maximum likelihood analysis is strongly influenced by the number of distinct site patterns in the concatenated alignment, whereas statistical power depends primarily on the amount of evolutionary information represented by the total number of sites and substitutions. By reducing the former while restoring the latter through upsampling, PSU can approximate many full-data analyses at substantially lower computational cost. Analysis of simulated and empirical datasets shows that PSU can accurately estimate bootstrap support values, select the optimal substitution model, test evolutionary hypotheses, and infer branch lengths, divergence times, and associated uncertainty measures, while reducing runtime and memory requirements by orders of magnitude. PSU also provides distributions of inferred clade support across independent subsamples, enabling detection of conflicting phylogenetic signals that may remain hidden in conventional bootstrap analysis. Automated tuning of subsample size, the number of subsamples, and the number of upsampling replicates make PSU practical across diverse datasets. We suggest that PSU is a general strategy for scalable phylogenomic inference using a broad range of statistical methods. By enabling analyses of genome-scale alignments on commodity hardware, PSU broadens research access and reduces environmental and infrastructural costs of big-data phylogenomics.

confidence limits↗
PubMed2003

Developing a prognostic model in the presence of missing data: an ovarian cancer case study.

When developing prognostic models in medicine, covariate data are often missing and the standard response is to exclude those individuals whose data are incomplete from the analyses. This practice leads to a reduction in the statistical power, and may lead to biased results. We wished to develop a prognostic model for overall survival from 1,189 primary cases (842 deaths) of epithelial ovarian cancer. A complete case analysis restricted the sample size to 518 (380 deaths). After applying a multiple imputation (MI) framework we included three real values for each one imputed, and constructed a model composed of more statistically significant prognostic factors and with increased predictive ability. Missing values can be imputed in cases where the reason for the data being missing is known, particularly where it can be explained by available data. This will increase the power of an analysis and may produce models that are more statistically reliable and applicable within clinical practice.

Adolescent↗
PubMedDate not supplied

Effects of narcotic and non-narcotic continuous epidural anesthesia on intrapartum fetal heart rate tracings as measured by computer analysis.

OBJECTIVE: To evaluate the effect of narcotic and non-narcotic continuous epidural anesthesia on intrapartum fetal heart rate (FHR) tracings as measured by computer analysis. METHODS: We studied 37 women with uncomplicated pregnancies at term with reactive FHR tracings. The women were randomized to receive epidural anesthesia with either bupivicaine with fentanyl or bupivicaine alone. One-hour FHR tracings were obtained before epidural anesthesia. Thirty minutes after the initial bolus of the epidural a repeat computer analysis of 60-minute FHR tracing was obtained. Median values are reported for FHR parameters with statistical analysis performed by the Mann-Whitney U and Wilcoxon signed rank tests where appropriate. A power calculation was performed using a power of 90% to determine a required sample size of 28 patients. Statistical significance was set at P < .05. RESULTS: In early first stage of labor, there was no significant difference in pre- and postepidural anesthesia FHR baseline, accelerations of 10 and 15 beats per minute, episodes of high and low variation, and short- and long-term variation when using either narcotic or non-narcotic anesthetic agents. CONCLUSIONS: Thus, the clinician can consider the use of narcotic as well as non-narcotic continuous epidural anesthesia in the dosages used in our study with its attendant advantages without fear of obscuring the fetal heart rate tracing.

Anesthesia, Epidural↗
PubMedDate not supplied

EEG correlates of neurotoxicity.

Electroencephalographic (EEG) techniques provide an interface between behavioral studies and more mechanistically directed neurophysiological studies, thus constituting a potentially important in vivo method of neurobehavioral assessment. The amplitude and frequency of the EEG signal depends on the behavioral state of the subject and the location of the recording electrode. Systematic assessment of EEG recordings may reveal alterations in integrated neuronal function induced by drug or neurotoxicant exposure. Fourier analysis allows more precise analysis of the frequency content of the recorded signal and yields data that may be further reduced and statistically analyzed. EEG data from neuropharmacologic and neurotoxicologic studies are discussed to illustrate application of power spectral analysis techniques in experimental laboratory settings.

Animals↗
PubMedDate not supplied

Effect of lateral positions on tissue oxygenation in the critically ill.

OBJECTIVE: The purpose of this study was to assess the effect of lateral positions on tissue oxygenation in critically ill patients. DESIGN: The study design was prospective and quasi-experimental, and we used a convenience sample with random assignment. SETTING: The study took place in the intensive care unit and the cardiac intensive care unit of a 450-bed medical center in the northwestern United States. PATIENTS: The sample included 12 adult patients with indwelling pulmonary artery and radial arterial catheters who were receiving mechanical ventilation and who met the criteria of "critical illness" by having impaired arterial oxygenation (PaO2 < or = 70 mm Hg) and/or cardiac index < or = 2.0 L/min/m2. OUTCOME MEASURES: The outcome measures were dependent variables reflecting oxygen delivery including heart rate, cardiac output, arterial oxygen content (CaO2) and oxygen consumption, and the adequacy of tissue oxygenation (serum lactate). INTERVENTION: Each patient was passively turned to each of the three positions (right and left 45 degrees lateral and supine) according to a computer-generated, randomized positioning sequence. Dependent variables were measured 15 minutes after each position change. No changes in ventilator settings or vasoactive drugs occurred during data collection. RESULTS: Analysis of variance for repeated measures was used in the data analysis. Post hoc analysis determined an effect size of 0.558 and power of 0.80 at an alpha level of.05. No statistically significant differences caused by position were found in mean CaO2, cardiac output, heart rate, respiratory rate, PaO2, SaO2, or lactate level. Pearson correlation analysis found no significant relationships between the primary variables reflecting oxygen delivery (cardiac output and CaO2) and serum lactate levels. CONCLUSIONS: These findings suggest that lateral positioning of critically ill patients who are hypoxemic or have low cardiac output does not further endanger tissue oxygenation. Evaluation of individual patient responses to position changes in the clinical setting is encouraged until further studies using more heterogenous populations can provide more definitive guidance.

Aged↗
PubMedDate not supplied

Combining independent studies of diagnostic fluorodeoxyglucose positron-emission tomography and computed tomography in mediastinal lymph node staging for non-small cell lung cancer.

AIMS AND BACKGROUND: A meta-analysis of diagnostic test performance was conducted to compare the results of relevant studies reporting diagnostic accuracy values for mediastinal staging in patients with non-small cell lung cancer (NSCLC). This paper deals with the two most accurate imaging techniques currently in use: positron emission tomography (PET) with FDG and computed tomography (CT). A statistical pooling method was used to perform a quantitative meta-analysis aimed at demonstrating the potential advantage of one of these two methods. METHODS: Studies in all languages published between 1998 and 2005 that examined the use of FDG-PET and CT for mediastinal staging in NSCLC patients, enrolled at least 18 participants, and provided enough data to allow calculation of sensitivity and specificity rates were considered eligible for the quantitative meta-analysis. Statistical methods to pool the overall estimates of sensitivity and specificity and to compare the discriminant power of PET and CT were discussed and used. RESULTS: Of the 13 studies included in the analysis, 12 reported greater accuracy of FDG-PET than CT in detecting mediastinal lymph node metastases. The sensitivity of FDG-PET ranged from 50% to 100%. The estimate of the overall sensitivity was 0.83% with 95% CI (0.749-0.913). Specificity ranged from 79% to 100%, with an overall estimated specificity of 0.87% with 95% CI (0.80-0.95). For CT, the sensitivity and specificity ranged from 50% to 97% and 58% to 94%, respectively; the overall estimate was 0.68% with 95% CI (0.582-0.788) and 0.76% with 95% CI (0.668-0.859). The summary receiver operating characteristic (SROC) approach was used to assess the superior diagnostic accuracy of one of the two methods. The areas under the two SROC curves were AUC(PET) = 0.909 vs. AUC(CT) = 0.794. CONCLUSIONS: Numerical and visual results of the meta-analysis of recent relevant reports agreed that FDG-PET is more accurate than CT in identifying mediastinal lymph node metastases in non-small cell lung cancer.

Aged↗
PubMed2006

The value of relatives with phenotypes but missing genotypes in association studies for quantitative traits.

The additional statistical power of association studies for quantitative traits was derived when ungenotyped relatives with phenotypes are included in the analysis. It was shown that the extra power is a simple function of the coefficient of additive genetic relationship and the phenotypic correlation coefficient between the genotyped and ungenotyped relatives. For close relatives, such as pairs of fullsibs and identical twin pairs, gains in power in the range of 10 to 30% are achieved if only one of the pair is genotyped. The theoretical results were verified by simulations. It was shown that ignoring the error in estimating the genotype of the ungenotyped relative has little impact on the estimates and on statistical power, consistent with results from quantitative trait loci (QTL) linkage studies. For genome-wide association studies in which not all relatives with phenotypes can be genotyped, our study provides a prediction of the additional power of an analysis that includes phenotypes on ungenotyped individuals, and can be used in experimental design. We show that a two-step procedure, in which missing genotypes are imputed and subsequently an association analysis is performed, is efficient and powerful.

Genetic Predisposition to Disease↗
PubMed2001

The power of the transmission disequilibrium test (TDT) with both case-parent and control-parent trios.

The transmission disequilibrium test (TDT) customarily uses affected children and their parents (often case-parent trios, TDTD). Control-parent trios are necessary to guard against spurious significant results due to segregation distortion but are not generally utilized in the identification of disease susceptibility loci (DSL). Controls are often easy to recruit and the TDT can easily be extended to include control-parent trios into the analyses with unrelated case-parent trios. We present an extension of the TDT (TDTDC) that incorporates unrelated cases and controls and their parents into a single analysis. We develop a simple and accurate analytical method for computing the statistical power of various TDT (e.g. the TDTD, TDTDC, TDTDC and TDTC that employ control-parent trios only) under any genetic model. We investigated the power of these TDT, and particularly compared the relative power of the TDTD and TDTDC. We found that the TDTDC is almost always more powerful than the TDTC and TDTD. The relative power of the TDTDC and TDTD depends largely upon a number of parameters identified in the study. This study provides a basis for efficient use of control-parent trios in DSL identification.

Case-Control Studies↗
PubMed1995

Is it necessary to transform nutrient variables prior to statistical analyses?

The distributions of the intakes of many nutrients are skewed, yet this is often overlooked when standard statistical analyses are applied to nutrient data. The nutrient intakes of 5,123 men and 5,236 women, recorded by food frequency questionnaire in the Scottish Heart Health Study, were transformed to achieve approximately symmetric distributions. Power transformations were chosen using letter value analyses. A letter value analysis uses selected order statistics and their position around the median to assess symmetry. The effect that each transformation had on a comparison of nutrient intakes between those with and without prevalent coronary heart disease was determined from t tests on the untransformed and transformed variable. The effects of the logarithm and square root transformation and of the optimum Box-Cox transformation were also determined, and the results were compared with the nonparametric Mann-Whitney test. The conclusion of whether or not to reject the null hypotheses often varied, depending on the transformation and test used. The nonparametric test usually gave a conclusion similar to that of the t test on the letter value-transformed data, the Box-Cox-transformed variable, and after either the logarithm or square root transformation of the data, but not always both. The results from the untransformed variable were sometimes very different. Failure to account for skewness in nutrient variables may thus lead to spurious conclusions.

Aged↗