SEARCH · PubMed Health

Results for “bootstrap”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 811 records · Page 45Linked to original sources

PubMed2003

Anaplasma marginale msp1alpha genotypes evolved under positive selection pressure but are not markers for geographic isolates.

Anaplasma marginale (order Rickettsiales, family Anaplasmataceae), a tick-borne pathogen of cattle, is endemic in tropical and subtropical regions of the world. Many geographic isolates of A. marginale occur in the United States and have been identified by major surface protein 1a (MSP1a), which varies in sequence and molecular weight due to different numbers of tandem 28- to 29-amino-acid repeats. The present study was undertaken to examine the genetic variations among isolates of A. marginale obtained during 2001 from infected cattle from east-central Oklahoma, where A. marginale is endemic. The gene and protein sequences of MSP1a and msp4 nucleotide sequences were used to infer the phylogenetic relationships among Oklahoma and New World isolates from Argentina, Brazil, Mexico, and the United States. All 11 A. marginale isolates collected from Oklahoma had different MSP1a sequences but identical MSP4 sequences. The phylogenies of the msp4 sequences of 13 isolates from Oklahoma in comparison with those of 7 Latin American isolates and 12 U.S. isolates by maximum-parsimony (MP) and maximum-likelihood (ML) analyses, with A. centrale and A. ovis sequences used as outgroups, provided strong bootstrap analysis support for a Latin American clade. Isolates of A. marginale from the southern United States (Florida, Mississippi, and Virginia) and the west-central United States (California, Idaho, Illinois, Oregon, Missouri, and Texas) also grouped into two clades. Both clades contained isolates from Oklahoma, suggesting extensive cattle movement. ML analysis of the msp4 sequences of isolates from Oklahoma provided bootstrap analysis support for east-central and north-central clades in Oklahoma, and both clades included isolates from Stillwater, Okla. Analysis of the codon and amino acid changes among the msp4 sequences of isolates with different phylogenies provided evidence that msp4 is not under positive selection pressure. In contrast, the phylogenies of the MSP1a DNA and protein sequences of 13 isolates from Oklahoma in comparison with those of 7 Latin American and 13 isolates from the United States by MP and ML analyses demonstrated no geographic clustering and provided evidence that this gene is under positive selection pressure. The results indicate that msp1alpha is not a marker for the characterization of A. marginale geographic isolates and suggest that the genetic heterogeneity observed among isolates of A. marginale within Oklahoma could be explained by cattle movement and the maintenance of different genotypes by independent transmission events.

Amino Acid Sequence↗

PubMed1997

Identification of GB virus C variants by phylogenetic analysis of 5'-untranslated and coding region sequences.

Phylogenetic analysis of 44 GB virus C (GBV-C) 5'-untranslated region (5'-UTR) sequences from 37 individuals suggested the presence of GBV-C genotypes (A. S. Muerhoff, J. N. Simons, T. P. Leary, J. C. Erker, M. L. Chalmers, T. J. Pilot-Matias, G. J. Dawson, S. M. Desai, and I. K. Mushahwar, J. Hepatol. 25:379-384, 1996) that correlated with geographic origin: type 1, 2a and 2b, and 3 isolates are found predominantly in West Africa, the United States and Europe, and Japan, respectively. We have extended our analysis to include 5'-UTR sequences from 129 globally distributed GBV-C isolates and sequences from the second envelope protein (E2) gene and nonstructural (NS) regions 3 and 5b from a subset of these isolates. Bootstrap analysis of a 157-nucleotide segment of the 5'-UTR from 129 sequences provided weak support for the existence of the four major groups of GBV-C isolates previously described, although phylogenetic analysis of a 374-nucleotide segment of the 5'-UTR from 83 isolates provided stronger support. Thus, the groups of GBV-C variants previously identified upon analysis of the entire 5'-UTR can be distinguished by analysis of the shorter, 374-nucleotide region from the 5'-UTR. In contrast, independent analysis of the E2, NS3, or NS5b region sequences does not identify groups of GBV-C variants that correlate with geographic origin. However, bootstrap analysis of these coding sequences, when linked to form colinear sequences, demonstrates that longer coding regions can produce GBV-C groupings that are similar to that determined from 5'-UTR sequence analysis. The inability to distinguish between GBV-C variants by using small segments of coding sequence suggests that the GBV-C genome is constrained. As a result of these constraints, there is a high degree of nucleotide and amino acid sequence conservation between isolates from widely separated geographic areas. Hence, substitutions at many nucleotide positions are not tolerated, so that substitutions at the positions which can change are saturated, thereby obscuring the evolutionary relationships.

Base Sequence↗

PubMed1999

Evidence of recombination among enteroviruses.

Human enteroviruses consist of more than 60 serotypes, reflecting a wide range of evolutionary divergence. They have been genetically classified into four clusters on the basis of sequence homology in the coding region of the single-stranded RNA genome. To explore further the genetic relationships between human enteroviruses and to characterize the evolutionary mechanisms responsible for variation, previously sequenced genomes were subjected to detailed comparison. Bootstrap and genetic similarity analyses were used to systematically scan the alignments of complete genomic sequences. Bootstrap analysis provided evidence from an early recombination event at the junction of the 5' noncoding and coding regions of the progenitors of the current clusters. Analysis within the genetic clusters indicated that enterovirus prototype strains include intraspecies recombinants. Recombination breakpoints were detected in all genomic regions except the capsid protein coding region. Our results suggest that recombination is a significant and relatively frequent mechanism in the evolution of enterovirus genomes.

Enterovirus↗

PubMed2006

A reliable phenotype predictor for human immunodeficiency virus type 1 subtype C based on envelope V3 sequences.

In human immunodeficiency virus type 1 (HIV-1) subtype B infections, the emergence of viruses able to use CXCR4 as a coreceptor is well documented and associated with accelerated CD4 decline and disease progression. However, in HIV-1 subtype C infections, responsible for more than 50% of global infections, CXCR4 usage is less common, even in individuals with advanced disease. A reliable phenotype prediction method based on genetic sequence analysis could provide a rapid and less expensive approach to identify possible CXCR4 variants and thus increase our understanding of subtype C coreceptor usage. For subtype B V3 loop sequences, genotypic predictors have been developed based on position-specific scoring matrices (PSSM). In this study, we apply this methodology to a training set of 279 subtype C sequences of known phenotypes (228 non-syncytium-inducing [NSI] CCR5(+) and 51 SI CXCR4(+) sequences) to derive a C-PSSM predictor. Specificity and sensitivity distributions were estimated by combining data set bootstrapping with leave-one-out cross-validation, with random sampling of single sequences from individuals on each bootstrap iteration. The C-PSSM had an estimated specificity of 94% (confidence interval [CI], 92% to 96%) and a sensitivity of 75% (CI, 68% to 82%), which is significantly more sensitive than predictions based on other methods, including a commonly used method based on the presence of positively charged residues (sensitivity, 47.8%). A specificity of 83% and a sensitivity of 83% were achieved with a validation set of 24 SI and 47 NSI unique subtype C sequences. The C-PSSM performs as well on subtype C V3 loops as existing subtype B-specific methods do on subtype B V3 loops. We present bioinformatic evidence that particular sites may influence coreceptor usage differently, depending on the subtype.

HIV Envelope Protein gp120↗

PubMed1997

Clinical trials in rheumatoid arthritis: methodological suggestions for assessing radiographs arising from the GRISAR Study. Gruppo Reumatologi Italiani Studio Artrite Reumatoide.

OBJECTIVES: The three x ray assessors of the GRISAR study (blinded to treatment) gave consensual erosion and damage scores to the baseline and 12 month radiographs of 284 rheumatoid arthritis (RA) patients using three different methods: single readings (blinded as to patient and chronological sequence of the x rays), paired readings (blinded as to sequence), and chronologically ordered paired readings. The aim was to evaluate which of these reading procedures is the most appropriate for clinical trials. METHODS: The progression of the scores obtained using each procedure was compared by means of descriptive statistics, principal components analysis, and intra-patient correlation coefficients of pairs of methods. Bootstrap estimates of the variance of the difference in the means of two equally sized random samples were calculated to evaluate the power of the statistical analysis performed to assess the possible treatment effect for both paired and chronological reading methods. RESULTS: (a) The standard deviations of the paired and chronological readings were similar, but that of the single readings was higher. (b) The knowledge that two x rays were of the same patient accounted for a sizeable proportion of the between method variability. (c) Agreement was satisfactory between the paired and chronological methods for both scores but, between them and the single readings, it was modest for erosions and poor for damage. (d) The bootstrap estimate of the variance of the difference was smaller for the paired than the chronological method, possibly giving it greater power to test treatment effect. CONCLUSIONS: These results suggested that paired readings were the most suitable for evaluating the progression of joint damage in the GRISAR study.

Arthritis, Rheumatoid↗

PubMed2001

Financial cost of social exclusion: follow up study of antisocial children into adulthood.

OBJECTIVES: To compare the cumulative costs of public services used through to adulthood by individuals with three levels of antisocial behaviour in childhood. DESIGN: Costs applied to data of 10 year old children from the inner London longitudinal study selectively followed up to adulthood. SETTING: Inner London borough. PARTICIPANTS: 142 individuals divided into three groups in childhood: no problems, conduct problems, and conduct disorder. MAIN OUTCOME MEASURES: Costs in 1998 prices for public services (excluding private, voluntary agency, indirect, and personal costs) used over and above basic universal provision. RESULTS: By age 28, costs for individuals with conduct disorder were 10.0 times higher than for those with no problems (95% confidence interval of bootstrap ratio 3.6 to 20.9) and 3.5 times higher than for those with conduct problems (1.7 to 6.2). Mean individual total costs were 70 019 pounds sterling for the conduct disorder group (bootstrap mean difference from no problem group 62 pound sterling; 898 pound sterling 22 692 pound sterling to 117 pound sterling) and 24 324 pound sterling (16 707 pound sterling; 6594 pound sterling to 28 149 pound sterling) for the conduct problem group, compared with 7423 pound sterling for the no problem group. In all groups crime incurred the greatest cost, followed by extra educational provision, foster and residential care, and state benefits; health costs were smaller. Parental social class had a relatively small effect on antisocial behaviour, and although substantial independent contributions came from being male, having a low reading age, and attending more than two primary schools, conduct disorder still predicted the greatest cost. CONCLUSIONS: Antisocial behaviour in childhood is a major predictor of how much an individual will cost society. The cost is large and falls on many agencies, yet few agencies contribute to prevention, which could be cost effective.

Child↗

PubMed2005

Multivariate prediction of major adverse cardiac events after 9914 percutaneous coronary interventions in the north west of England.

OBJECTIVE: To develop a multivariate prediction model for major adverse cardiac events (MACE) after percutaneous coronary interventions (PCIs) by using the North West Quality Improvement Programme in Cardiac Interventions (NWQIP) PCI Registry. SETTING: All NHS centres undertaking adult PCIs in north west England. METHODS: Retrospective analysis of prospectively collected data on 9914 consecutive patients undergoing adult PCI between 1 August 2001 and 31 December 2003. A multivariate logistic regression analysis was undertaken, with the forward stepwise technique, to identify independent risk factors for MACE. The area under the receiver operating characteristic (ROC) curve and the Hosmer-Lemeshow goodness of fit statistic were calculated to assess the performance and calibration of the model, respectively. The statistical model was internally validated by using the technique of bootstrap resampling. MAIN OUTCOME MEASURES: MACE, which were in-hospital mortality, Q wave myocardial infarction, emergency coronary artery bypass graft surgery, and cerebrovascular accidents. RESULTS: Independent variables identified with an increased risk of developing MACE were advanced age, female sex, cerebrovascular disease, cardiogenic shock, priority, and treatment of the left main stem or graft lesions during PCI. The ROC curve for the predicted probability of MACE was 0.76, indicating a good discrimination power. The prediction equation was well calibrated, predicting well at all levels of risk. Bootstrapping showed that estimates were stable. CONCLUSIONS: A contemporaneous multivariate prediction model for MACE after PCI was developed. The NWQIP tool allows calculation of the risk of MACE permitting meaningful risk adjusted comparisons of performance between hospitals and operators.

Adult↗

PubMed2005

Trial-to-trial variability and its effect on time-varying dependency between two neurons.

The joint peristimulus time histogram (JPSTH) and cross-correlogram provide a visual representation of correlated activity for a pair of neurons, and the way this activity may increase or decrease over time. In a companion paper we showed how a Bootstrap evaluation of the peaks in the smoothed diagonals of the JPSTH may be used to establish the likely validity of apparent time-varying correlation. As noted in earlier studies by Brody and Ben-Shaul et al., trial-to-trial variation can confound correlation and synchrony effects. In this paper we elaborate on that observation, and present a method of estimating the time-dependent trial-to-trial variation in spike trains that may exceed the natural variation displayed by Poisson and non-Poisson point processes. The statistical problem is somewhat subtle because relatively few spikes per trial are available for estimating a firing-rate function that fluctuates over time. The method developed here decomposes the spike-train variability into a stimulus-related component and a trial-specific component, allowing many degrees of freedom to characterize the former while assuming a small number suffices to characterize the latter. The Bootstrap significance test of the companion paper is then modified to accommodate these general excitability effects. This methodology allows an investigator to assess whether excitability effects are constant or time-varying, and whether they are shared by two neurons. In data from two V1 neurons we find that highly statistically significant evidence of dependency disappears after adjustment for time-varying trial-to-trial variation.

Action Potentials↗

PubMed2005

Statistical assessment of time-varying dependency between two neurons.

The joint peristimulus time histogram (JPSTH) provides a visual representation of the dynamics of correlated activity for a pair of neurons. There are many ways to adjust the JPSTH for the time-varying firing-rate modulation of each neuron, and then to define a suitable measure of time-varying correlated activity. Our approach is to introduce a statistical model for the time-varying joint spiking activity so that the joint firing rate can be estimated more efficiently. We have applied an adaptive smoothing method, which has been shown to be effective in capturing sudden changes in firing rate, to the ratio of joint firing probability to the probability of firing predicted by independence. A bootstrap procedure, applicable to both Poisson and non-Poisson data, was used to define a statistical significance test of whether a large ratio could be attributable to chance alone. A numerical simulation showed that the bootstrap-based significance test has very nearly the correct rejection probability, and can have markedly better power to detect departures from independence than does an approach based on testing contiguous bins in the JPSTH. In a companion paper, we show how this formulation can accommodate latency and time-varying excitability effects, which can confound spike timing effects.

Action Potentials↗

PubMed2005

Gene expression profiling allows distinction between primary and metastatic squamous cell carcinomas in the lung.

Lung neoplasms commonly develop in patients previously treated for head and neck carcinomas. The derivation of these tumors, either as new primary lung cancers or as metastatic head and neck cancers, is difficult to establish based on clinical or histopathologic criteria since both are squamous cell carcinomas and have identical features under light microscopy. However, this distinction has significant treatment and prognostic implications. Gene expression profiling was performed on a panel of 52 sequentially collected patients with either primary lung (n = 21) or primary head and neck (n = 31) carcinomas using the Affymetrix HG_U95Av2 high-density oligonucleotide microarray. Unsupervised hierarchical clustering with Ward linkage and the Pearson correlation metric was performed. To assess robustness, bootstrap resampling was performed with 1,000 iterations. A t test of the normalized values for each gene was used to determine the genes responsible for segregating head and neck from lung primary carcinomas, and those with the most differential expression were used for later analyses. In the absence of a large "test" set of tumors, we used a supervised leave-one-out cross-validation to test how well we could predict the tumor origin. Once a gene expression profile was established, 12 lung lesions taken from patients with previously treated head and neck cancers were similarly analyzed by gene expression profiling to determine their sites of origin. Unsupervised clustering analysis separated the study cohort into two distinct groups which reliably remained segregated with bootstrap resampling. Group 1 consisted of 30 tongue carcinomas. Group 2 consisted of 21 lung cancers and 1 tongue carcinoma. The clustering was not changed even when normal lung or tongue profiles were subtracted from the corresponding carcinomatous lesions, and a leave-one-out cross-validation showed a 98% correct prediction (see Supplementary Data 1). A minimum set of 500 genes required to distinguish these groups was established. Given the ability to segregate these lesions using molecular profiling, we analyzed the lung tumors of undetermined origin. All cases clearly clustered with either lung or tongue tumor subsets, strongly supporting our hypothesis that this technique could elucidate the tissue of origin of metastatic lesions. Although histologically similar, squamous cell carcinomas have distinct gene expression profiles based on their anatomic sites of origin. Accordingly, the application of gene expression profiling may be useful in identifying the derivation of lung nodules and consequently enhances treatment planning.

Algorithms↗

PubMed2005

Cost-effectiveness of dual-chamber pacing compared with ventricular pacing for sinus node dysfunction.

BACKGROUND: Compared with single-chamber ventricular pacing, dual-chamber pacing can reduce adverse events and, as a result, improve quality of life in patients paced for sick sinus syndrome. It is not clear, however, how these benefits compare with the increased cost of dual-chamber pacemakers. METHODS AND RESULTS: We used 4-year data from a 2010-patient, randomized trial to estimate the incremental cost-effectiveness of dual-chamber pacing compared with ventricular pacing and then projected these findings over the patients' lifetimes by using a Markov model that was calibrated to the first 5 years of in-trial data. To assess the stability of the findings, we performed 1000 bootstrap analyses and multiple sensitivity analyses. During the first 4 years of the trial, dual-chamber pacemakers increased quality-adjusted life expectancy by 0.013 year per subject at an incremental cost-effectiveness ratio of 53,000 dollars per quality-adjusted year of life gained. Over a lifetime, dual-chamber pacing was projected to increase quality-adjusted life expectancy by 0.14 year with an incremental cost-effectiveness ratio of approximately 6800 dollars per quality-adjusted year of life gained. In bootstrap analyses, dual-chamber pacing was cost-effective in 91.9% of simulations at a threshold of 50,000 dollars per quality-adjusted year of life and in 93.2% of simulations at a threshold of 100,000 dollars. Its cost-effectiveness ratio was also below this threshold in numerous sensitivity analyses that varied key estimates. CONCLUSIONS: For patients with sick sinus syndrome requiring pacing, dual-chamber pacing increases quality-adjusted life expectancy at a cost that is generally considered acceptable.

Aged↗

PubMed2000

Progressive resistance exercise and resting blood pressure : A meta-analysis of randomized controlled trials.

Hypertension is a major public health problem affecting an estimated 43 million civilian, noninstitutionalized adults in the United States (24% of this population). The purpose of this study was to use the meta-analytic approach to examine the effects of progressive resistance exercise on resting systolic and diastolic blood pressure in adult humans. Studies were retrieved via (1) computerized literature searches, (2) cross-referencing from original and review articles, and (3) review of the reference list by 2 experts on exercise and blood pressure. Inclusion criteria were as follows: (1) trials that included a randomized nonexercise control group; (2) progressive resistance exercise as the only intervention; (3) adult humans; (4) journal articles, dissertations, and masters theses published in the English-language literature; (5) studies published and indexed between January 1966 and December 1998; (6) resting systolic and/or diastolic blood pressure assessed; and (7) training studies lasting a minimum of 4 weeks. Across all designs and categories, fixed-effects modeling yielded decreases of approximately 2% and 4% for resting systolic and diastolic blood pressure, respectively (mean+/-SD systolic, -3+/-3 mm Hg; 95% bootstrap CI, -4 to -1 mm Hg; mean+/-SD diastolic, -3+/-2 mm Hg; 95% bootstrap CI, -4 to -1 mm Hg). It was concluded that progressive resistance exercise is efficacious for reducing resting systolic and diastolic blood pressure in adults. However, a need exists for additional studies that limit enrollment to hypertensive subjects as well as analysis of data with an intention-to-treat approach before the effectiveness of progressive resistance exercise as a nonpharmacological intervention can be determined.

Adult↗

PubMed2006

Can multivariable risk-benefit profiling be used to select treatment-favorable patients for thrombolysis in stroke in the 3- to 6-hour time window?

BACKGROUND AND PURPOSE: The Stroke-Thrombolytic Predictive Instrument (Stroke-TPI) uses multivariate equations to predict outcomes with and without thrombolysis. We sought to examine whether such a multivariate predictive instrument might be useful in selecting patients with a favorable risk-benefit treatment profile for therapy after 3 hours. METHODS: We explored outcomes in patients from 5 major randomized clinical trials testing intravenous recombinant tissue plasminogen activator (rt-PA) classified by the Stroke-TPI as "treatment-favorable" or "treatment-unfavorable." We used iterative bootstrap re-sampling to estimate how such a model would perform on independent test data. RESULTS: Among patients treated within the 3- to 6-hour window, 67% of patients were classified by Stroke-TPI predicted outcomes as "treatment-favorable" and 33% were classified as "treatment-unfavorable." Outcomes in the treatment-favorable group demonstrated benefit for thrombolysis (modified Rankin Scale score < or =1: 44.0% with rt-PA versus 34.2 with placebo, P=0.005), whereas harm was demonstrated in the treatment-unfavorable group (modified Rankin Scale score < or =1: 31.3% with rt-PA versus 38.3% with placebo; P=0.004). Bootstrap resampling with complete cross-validation showed that the absolute margin of benefit in the treatment-favorable group diminished on average by 36% between derivation and independent validation sets, but still represented a significant tripling of improvement in benefit compared with conventional inclusion criteria (5.2% [interquartile range, 1.7% to 8.6%] versus 1.8% [interquartile range, -0.5 to 4.1], P<0.0001). CONCLUSIONS: Such multivariable risk-benefit profiling may be useful in the selection of acute stroke patients for rt-PA therapy even more than 3 hours after symptom onset. Prospective testing is indicated.

Aged↗

PubMed2001

Chiropractic manipulation and stroke: a population-based case-control study.

BACKGROUND AND PURPOSE: Several reports have linked chiropractic manipulation of the neck to dissection or occlusion of the vertebral artery. However, previous studies linking such strokes to neck manipulation consist primarily of uncontrolled case series. We designed a population-based nested case-control study to test the association. METHODS: Hospitalization records were used to identify vertebrobasilar accidents (VBAs) in Ontario, Canada, during 1993-1998. Each of 582 cases was age and sex matched to 4 controls from the Ontario population with no history of stroke at the event date. Public health insurance billing records were used to document use of chiropractic services before the event date. RESULTS: Results for those aged <45 years showed VBA cases to be 5 times more likely than controls to have visited a chiropractor within 1 week of the VBA (95% CI from bootstrapping, 1.32 to 43.87). Additionally, in the younger age group, cases were 5 times as likely to have had >/=3 visits with a cervical diagnosis in the month before the case's VBA date (95% CI from bootstrapping, 1.34 to 18.57). No significant associations were found for those aged >/=45 years. CONCLUSIONS: While our analysis is consistent with a positive association in young adults, potential sources of bias are also discussed. The rarity of VBAs makes this association difficult to study despite high volumes of chiropractic treatment. Because of the popularity of spinal manipulation, high-quality research on both its risks and benefits is recommended.

Adult↗

PubMed2005

Oxygenation index predicts outcome in children with acute hypoxemic respiratory failure.

To define outcome and time dependence of predictors of outcome in pediatric acute hypoxemic respiratory failure, 131 patients (age range, 1 month to 18 years) were prospectively followed. Parametric models were used to describe time-related events, and competing risks analysis was performed for mortality estimates. Multiple logistic analysis was applied to describe time-related predictors of ventilation time and mortality. Overall mortality was 27%. Peak oxygenation index (OI) measured at any time point (p < 0.001, 91% reliability in bootstrapping, after inverse transformation) and Pediatric Risk of Mortality, or PRISM, score within the first 12 hours of mechanical ventilation (p < 0.001, 63% reliability in bootstrapping, after square transformation) were identified as independent predictors of mortality. Peak OI, younger age, and need for renal replacement therapy were significantly associated with a longer time to extubation. Although OI was less reliable as outcome predictor within the first 12 hours of intubation, it still predicted duration of mechanical ventilation. No clear-cut threshold of OI was identified that could accurately predict mortality. Survival was characterized by a peak rate of extubations at approximately 1 week, with a more gradual decline thereafter, whereas death appeared as a constant risk over time, which exceeded chances of survival at approximately 4 weeks. Severity of oxygenation failure at any point in time during acute hypoxemic respiratory failure correlates with duration of mechanical ventilation and mortality. This is best reflected by the OI, which shows a direct correlation to outcome in a time-independent manner.

Acute Disease↗

PubMedDate not supplied

A note on the estimation of confidence intervals for cost-effectiveness when costs and effects are censored.

BACKGROUND: The relation between methodological advances in estimation of confidence intervals (CIs) for incremental cost-effectiveness ratios (ICER) and estimation of cost effectiveness in the presence of censoring has not been explored. The authors address the joint problem of estimating ICER precision in the presence of censoring. METHODS: Using patient-level data (n = 168) on cost and survival from a published placebo-controlled trial, the authors compared 2 methods of measuring uncertainty with censored data: 1) Bootstrap with censor adjustment (BCA); 2) Fieller's method with censor adjustment (FCA). The authors estimate the FCA over all possible values for the correlation (rho) between costs and effects (range = -1 to + 1) and also examine the use of the correlation between cases without censoring adjustment (i.e., simple time-on-study) for costs and effects as an approximation for rho. RESULTS: Using time-on-study, which considers all censored observations as responders (deaths), yields 0.64 life-years gained at an additional cost of 87.9 for a cost per life-year of 137 (95% CI by bootstrap -5.9 to 392). Censoring adjustment corrects for the bias in the time-on-study approach and reduces the cost per life-year estimate to 132 (=72/0.54). Confidence intervals with censor adjustment were approximately 40% wider than the base-case without adjustment. Using the Fieller method with an approximation of rho based on the uncensored cost and effect correlation provides a 95% CI of (-48 to 529), which is very close to the BCA interval of (-52 to 504). CONCLUSIONS: Adjustment for censoring is necessary in cost-effectiveness studies to obtain unbiased estimates of ICER with appropriate uncertainty limits. In this study, BCA and FCA methods, the latter with approximated covariance, are simple to compute and give similar confidence intervals.

Confidence Intervals↗

PubMedDate not supplied

Resampling techniques in the analysis of non-binormal ROC data.

The methods most commonly used for analyzing receiver operating characteristic (ROC) data incorporate "binormal" assumptions about the latent frequency distributions of test results. Although these assumptions have proved robust to a wide variety of actual frequency distributions, some data sets do not "fit" the binormal model. In such cases, resampling techniques such as the jackknife and the bootstrap provide versatile, distribution-independent, and more appropriate methods for hypothesis testing. This article describes the application of resampling techniques to ROC data for which the binormal assumptions are not appropriate, and suggests that the bootstrap may be especially helpful in determining confidence intervals from small data samples. The widespread availability of ever-faster computers has made resampling methods increasingly accessible and convenient tools for data analysis.

Bias↗

PubMed2006

Factors associated with the effect-size of thiazolidinedione (TZD) therapy on HbA(1c): a meta-analysis of published randomized clinical trials.

OBJECTIVE: To examine factors affecting the size of the HbA(1c) response to thiazolidinedione (TZD) therapy. RESEARCH DESIGN AND METHODS: Meta-analysis of randomized TZD controlled trials which were identified using PubMed, EBSCO and Sci-lit databases and were published in English. Sociodemographic and clinical data were extracted from each trial. HbA(1c) effect size was defined as either a placebo-subtracted change in HbA(1c) or a change in HbA(1c) from baseline. Weighted multivariable regression was used to examine factors associated with changes in HbA(1c). Bootstrapped smearing estimates were computed to obtain reliable estimates of HbA(1c) effect size. RESULTS: Forty-two trials yielded 60 trial arms which represented 8322 patients treated with thiazolidinediones. Weighted placebo-subtracted change in HbA(1c) was -0.99% +/- 0.02% with an average baseline HbA(1c) of 9.1% +/- 1.0%. Weighted bootstrapped smearing estimate of the placebo-subtracted change in HbA(1c) was -1.02% +/- 0.004%. After controlling for other variables, the baseline HbA(1c) level had a significant negative association with placebo-subtracted HbA(1c) change (p = 0.004) and also with change in HbA(1c) from baseline (p = 0.002). Longer trial duration was associated with greater placebo-subtracted HbA(1c) change (p = 0.01) but not with the change in HbA(1c) from baseline. The multivariable models explained 72% of the variation in placebo-subtracted HbA(1c) change. It was not possible to estimate effects of the run-in period and obesity on TZD effect size. CONCLUSION: Baseline HbA(1c) and trial duration significantly impacted the effect size of TZD therapy on HbA(1c). Age, gender, duration of diabetes and prior use of anti-diabetic therapy were not associated with the TZD effect size.

Diabetes Mellitus↗