Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Sample size estimation”

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 703 records · Page 39Linked to original sources
PubMed1999

Misclassification in case-control studies of gene-environment interactions: assessment of bias and sample size.

In studies of gene-environment interactions, exposure misclassification can lead to bias in the estimation of an interaction effect and increased sample size. The magnitude of the bias and the consequent increase in sample size for fixed misclassification probabilities are highly dependent on the prevalence of the misclassified factor and on the interaction model. This paper describes a relatively simple approach to assess the impact of misclassification on bias in the estimation of multiplicative or additive interactions and on sample size requirements. Applications of this method illustrate that even small errors in the assessment of environmental or genetic factors can result in biased interaction parameters and substantially increased sample size requirements that can compromise the feasibility of the study. Also, an example is provided where nondifferential misclassification biases an additive interaction parameter away from the null value, even under conditions where a multiplicative interaction parameter will always be biased toward the null value. Efforts to improve the accuracy in measuring both genetic and environmental factors are critical for the valid assessment of gene-environment interactions in case-control studies.

Benzo(a)pyrene↗
PubMed2000

Attack rates of human papillomavirus type 16 and cervical neoplasia in primiparous women and field trial designs for HPV16 vaccination.

BACKGROUND: Identification of human papillomavirus type 16 (HPV16) as the major risk factor for cervical neoplasia, and mass production of DNA free HPV capsids have paved the way to preventive vaccination trials. Design of such trials requires reliable attack rate data. OBJECTIVE: Determination of (1) HPV16 and (2) cervical neoplasia attack rates in primiparous women. Estimation of actuarial sample sizes for HPV16 vaccination phase IV trials. DESIGN: A longitudinal cohort study. METHODS: Population based Finnish Maternity Cohort (FMC) and Finnish Cancer Registry (FCR) were linked for the identification of two cohorts of primiparous women: (1) a random subsample of the FMC: 1656 women with two pregnancies between 1983-9 or 1990-6 and living in the Helsinki metropolitan area, and (2) all 72,791 primiparous women living in the same area during 1983-94. Attack rate for persistent HPV16 infection (1) was estimated in 1279 seronegative women by proportion of seroconversions between the first and the second pregnancy. Comparable 10 year cumulative incidence rate (CR) of cervical intraepithelial neoplasia grade III and cervical cancer (CIN III+) (2) was estimated based on cases registered at the FCR during 1991-4. RESULTS: The HPV16 attack rates were 13.8% (< 18 years), 7.0% (18-19 years), 2.3% (21 years), 2.4% (23 years), and 4.5% (< 25 years). Number of vaccinees required for a 5 year efficacy trial with persistent HPV16 infection as the end point ranged between 1000 and 3900, assuming 80% power, 90%-70% vaccine efficacy (VE), and misclassification. The CRs of CIN III+ were 0.33% (< 18 years), 0.44% (18-19 years), 0.21% (20-24 years), and 0.28% (< 25 years). Number of vaccinees required for a 10 year efficacy trial with HPV16 positive CIN III+ as the end point was 15,000 assuming 80% power, 90% VE, and 75% aetiological fraction of CIN III+ for HPV16. CONCLUSIONS: The attack rates of HPV16 and CIN III+ identify primiparous women under 25 years of age among target populations for postnatal HPV vaccination at phase II/III trials.

Adolescent↗
PubMed2004

Power of microsatellite markers for fingerprinting and parentage analysis in Eucalyptus grandis breeding populations.

We report the genetic analysis of 192 unrelated individuals of an elite breeding population of Eucalyptus grandis (Hill ex Maiden) with a selected set of six highly polymorphic microsatellite markers developed for species of the genus Eucalyptus. A full characterization of this set of six loci was carried out generating allele frequency distributions that were used to estimate parameters of genetic information content of these loci, including expected heterozygosity, polymorphism information content (PIC), power of exclusion, and probability of identity. The number of detected alleles per locus ranged from 6 to 33, with an average of 19.8 +/- 9.2. The average expected heterozygosity was 0.86 +/- 0.11 and the average PIC was 0.83 +/- 0.16. Using only three loci, it was possible to discriminate all 192 individuals. The overall probability of identity considering all six EMBRA microsatellite markers combined was lower than 1 in 2 billion. An analysis of the sample size necessary to estimate expected heterozygosity with minimum variance indicated that at least 64 individuals have to be genotyped to characterize this parameter with adequate accuracy for most microsatellites in Eucalyptus. The high degree of multiallelism and the clear and simple codominant Mendelian inheritance of the set of microsatellites used provide an extremely powerful system for the unique identification of Eucalyptus individuals for fingerprinting purposes and parentage testing.

Brazil↗
PubMed2005

Clozapine-induced weight gain: a study in monozygotic twins and same-sex sib pairs.

To assess the relative contribution of genetic factors in antipsychotic-induced weight gain, we explored the similarity in body mass index (BMI) (kg/m(2)) change under clozapine only (clozapine DeltaBMI) and upon additional inclusion of BMI change under prior antipsychotic medication (total DeltaBMI) of five monozygotic twins in comparison with seven same-sex sibs. Twin and sib pairs were identified by a telephone screening of 786 office-based psychiatrists. Measured data on weight and other clinical variables were obtained cross-sectionally and retrospectively from medical records. We found greater similarity in total DeltaBMI in monozygotic twins (intrapair difference 2.78+/-3.41 kg/m(2)) than in same-sex sibs (5.55+/-4.35 kg/m(2)), resulting in heritability estimates of h(2)=0.8 and A=0.45 (ACE twin model). However, intrapair differences in clozapine DeltaBMI were similar between twins (4.18+/-4.27 kg/m(2)) and sibs (4.68+/-4.88 kg/m(2)). We hypothesize that the weight plateau achieved under clozapine is influenced by genetic factors. The weight gain achieved during pretreatment with other antipsychotics seems to limit clozapine-induced weight gain, thus presumably explaining why heritability/similarity in monozygotic twins in comparison with same-sex sibs is greater for total DeltaBMI than for clozapine DeltaBMI. An important caveat is that, owing to the sample size, the heritability estimates have a large standard error and thus have to be interpreted with caution.

Adolescent↗
PubMed2006

The reassessment of trial perspectives from interim data--a critical view.

If an interim analysis is performed during a trial it is tempting to determine the conditional power to reach a rejection in the trial given the observed results in the interim analysis. Since the true effect size is unknown the conditional power may be calculated by using the effect size, which the study has been powered for in the planning phase or by using an interim estimate of the true size (or a combination of both). In either case the conditional power is a random variable and its density is investigated depending on the analysis time and the true effect size. Under the null hypothesis, in early interim analyses after a small proportion of sample units, the conditional power typically will be close to the overall power when the effect size from the planning stage is used for calculation. In this case the majority of observations must still be made and the small first-stage sample in general will be dominated by the hypothetical second-stage chance based on the wrong parameter value. It is shown that the conditional power in moderately underpowered studies can have a distribution symmetric around 0.5. When using the interim estimate for calculating the conditional power the density in general will be u-shaped. The impact of using conditional power to reassess the sample size using flexible two-stage combination tests is shown for a specific example in terms of overall power and average sample size as compared to the corresponding group sequential design. For small true effect sizes mid-trial sample size recalculation based on an interim estimate may lead to an overly large price to be paid in average sample size in relation to the gain in overall power. Finally, the problem is discussed in terms of estimating the true conditional power.

Clinical Trials as Topic↗
PubMed2004

A comparison of different strategies for computing confidence intervals of the linkage disequilibrium measure D'.

Many linkage disequilibrium (LD) measures have been used to study LD patterns and for haplotype block partitioning. We examine the properties of one of these measures, Lewontin's D', in order to understand the dependency of its confidence interval (CI) to allele frequency and sample size as well as its applications in defining haplotype blocks. This measure and its CIs were used to partition haplotypes into blocks by Gabriel et al. as well as in many other applications. Gabriel et al. utilized a bootstrap approach to calculate the CI for D'. Under this method, over 1,000 bootstrap samples may be needed to obtain an accurate estimate of the CI for each pair of single nucleotide polymorphism (SNP) markers which can be very computationally intensive, particularly when many SNP markers are involved. We develop two alternative methods for calculating the CI for D' without bootstrap: one based on the approximate variance of D' given by Zapata et al. and the other based on a maximum likelihood estimate (MLE) of D' together with Fisher Information theory. Both methods depend on normal approximation for the estimates of D' for large sample sizes. We assess and compare the coverage of the CIs using the three methods through extensive simulations. We define the coverage as the fraction of times the estimated CI contains the true value of D'. In general, the average coverage of the bootstrap method is less than the pre-specified coverage. When the sample size is small (< or = 100), the remaining two methods slightly under estimate the coverage with MLE approach having smaller standard error compared to Zapata's method. When the sample size is large (> or = 200) , the estimated coverage from both Zapata's and MLE methods are very close to the pre-specified coverage with the MLE method having the smallest standard error among all three methods. In most typical scenarios, we recommend the use of MLE method for all sample sizes. Only under rare specific cases, would the bootstrap method be better suited for determining the CI, i.e. small sample size, at extreme allele frequencies and -3 < D' < 0.

Computational Biology↗
PubMedDate not supplied

Adaptive designs for pivotal trials: discussion points from the PSI Adaptive Design Expert Group.

The Committee for Medicinal Products for Human Use (CHMP) is currently preparing a guideline on 'methodological issues in confirmatory clinical trials with flexible design and analysis plan'. PSI (Statisticians in the Pharmaceutical Industry) sponsored a meeting of pharmaceutical statisticians with an interest in the area to share experiences and identify potential opportunities for adaptive designs in late-phase clinical drug development. This article outlines the issues raised, resulting discussions and consensus views reached. Adaptive designs have potential utility in late-phase clinical development. Sample size re-estimation seems to be valuable and widely accepted, but should be made independent of the observed treatment effect where possible. Where unblinding is necessary, careful consideration needs to be given to preserving the integrity of the trial. An area where adaptive designs can be particularly beneficial is to allow dose selection in pivotal trials via adding/dropping treatment arms; for example, combining phase II and III of the drug development program. The more adaptations made during a late-phase clinical trial, the less likely that the clinical trial would be considered as a confirmatory trial. In all cases it would be advisable to consult with regulatory agencies at the protocol design stage. All involved should remain open to scientifically valid opportunities to improve drug development.

Clinical Trials as Topic↗
PubMed1994

Diagnostic performance of magnetic resonance imaging for the diagnosis of rotator cuff tears using supplemental images in the oblique sagittal plane.

RATIONALE AND OBJECTIVES: The authors evaluated the diagnostic utility of supplemental imaging in the oblique sagittal (OS) plane for the magnetic resonance imaging (MRI) diagnosis of rotator cuff tears. METHODS: Two radiologists with varying levels of MRI experience blindly reviewed shoulder MR examinations of 50 patients for rotator cuff tears. Shoulder examinations were interpreted twice, initially using only double-echo spin-echo images obtained in the oblique coronal (OC) plane and later using double-echo spin-echo images obtained in the both the OC and OS planes. Tears were characterized according to size, location, and extent, and levels of diagnostic confidence were evaluated. RESULTS: Sensitivity and specificity of MR for the diagnosis of rotator cuff tear was 85% and 80%, respectively, with the OC series increasing to 95% and 93%, respectively with OCOS scans. However, these increases were not statistically significant. Receiver operating characteristic (ROC) curves suggest a trend toward increased diagnostic confidence when supplemental OS scans are available, especially for the less experienced reader. Characterization of rotator cuff tears was not improved with additional OS images. CONCLUSIONS: Supplemental OS scans did not demonstrate a definite improvement in diagnostic accuracy for rotator cuff tears compared to OC scans alone. Estimations based on sample size calculations indicate that a much larger population of patients would be needed to show a statistically significant difference.

Humans↗
PubMed2005

The use of weighted Z-tests in medical research.

Traditionally the un-weighted Z-tests, which follow the one-patient-one-vote principle, are standard for comparisons of treatment effects. We discuss two types of weighted Z-tests in this manuscript to incorporate data collected in two (or more) stages or in two (or more) regions. We use the type A weighted Z-test to exemplify the variance spending approach in the first part of this manuscript. This approach has been applied to sample size re-estimation. In the second part of the manuscript, we introduce the type B weighted Z-tests and apply them to the design of bridging studies. The weights in the type A weighted Z-tests are pre-determined, independent of the prior observed data, and controls alpha at the desired level. To the contrary, the weights in the type B weighted Z-tests may depend on the prior observed data; and the type I error rate for the bridging study is usually inflated to a level higher than that of a full-scale study. The choice of the weights provides a simple statistical framework for communication between the regulatory agency and the sponsor. The negotiation process may involve practical constrains and some characteristics of prior studies.

Algorithms↗
PubMed2004

Risk adjustment effect on stroke clinical trials.

BACKGROUND AND PURPOSE: The ischemic stroke population is heterogeneous. Even in balanced randomized trials, patient heterogeneity biases estimates of the treatment effect toward no effect when dichotomous end points are used. Risk adjustment statistically addresses some of the heterogeneity and can reduce bias in the treatment effect estimate. The purpose of this study was to estimate the treatment effect of tissue plasminogen activator (tPA) in the National Institute of Neurological Disorders and Stroke (NINDS) tPA data set with and without adjustment for baseline differences. METHODS: Using a prespecified predictive model, we calculated unadjusted and risk-adjusted odds ratios (ORs) for favorable outcome for the Barthel Index, National Institutes of Health Stroke Scale, and Glasgow Outcome Scale for the patients in the NINDS tPA stroke trial. To assess the importance of the difference, a new sample size was calculated through the use of the risk-adjusted analysis. RESULTS: We analyzed 615 subjects. The ORs for the Barthel Index were 1.76 (unadjusted) and 2.04 (adjusted). The National Institutes of Health Stroke Scale and Glasgow Outcome Scale analyses also demonstrated increased ORs after adjustment. The estimated sample size required for the adjusted comparison was 13% smaller than the unadjusted sample. CONCLUSIONS: Risk adjustment in this data set suggests that the true treatment effect was larger than estimated by the unadjusted analysis. Stroke clinical trials should include prospective risk adjustment methodologies.

Aged↗
PubMed2005

Accuracy in the estimation of quantitative minimal area from the diversity/area curve.

The problem of representativity is fundamental in ecological studies. A qualitative minimal area that gives a good representation of species pool [C.M. Bouderesque, Methodes d'etude qualitative et quantitative du benthos (en particulier du phytobenthos), Tethys 3(1) (1971) 79] can be discerned from a quantitative minimal area which reflects the structural complexity of community [F.X. Niell, Sobre la biologia de Ascophyllum nosodum (L.) Le Jolis en Galicia, Invest. Pesq. 43 (1979) 501]. This suggests that the populational diversity can be considered as the value of the horizontal asymptote corresponding to the curve sample diversity/biomass [F.X. Niell, Les applications de l'index de Shannon a l'etude de la vegetation interdidale, Soc. Phycol. Fr. Bull. 19 (1974) 238]. In this study we develop a expression to determine minimal areas and use it to obtain certain information about the community structure based on diversity/area curve graphs. This expression is based on the functional relationship between the expected value of the diversity and the sample size used to estimate it. In order to establish the quality of the estimation process, we obtained the confidence intervals as a particularization of the functional (h-phi)-entropies proposed in [M. Salicru, M.L. Menendez, D. Morales, L. Pardo, Asymptotic distribution of (h,phi)-entropies, Commun. Stat. (Theory Methods) 22 (7) (1993) 2015]. As an example used to demonstrate the possibilities of this method, and only for illustrative purposes, data about a study on the rocky intertidal seawed populations in the Ria of Vigo (N.W. Spain) are analyzed [F.X. Niell, Estudios sobre la estructura, dinamica y produccion del Fitobentos intermareal (Facies rocosa) de la Ria de Vigo. Ph.D. Mem. University of Barcelona, Barcelona, 1979].

Algorithms↗
PubMedDate not supplied

The logistic analysis of epidemiologic prospective studies: investigation by simulation.

We performed a Monte Carlo computer simulation of the Walker-Duncan logistic regression technique in a typical epidemiologic prospective setting and analysed the results with respect to the accuracy and reliability of the regression estimates and the associated statistical significance tests (Z-tests). The results strongly suggest that the estimates were neither accurate nor reliable. The magnitude of the difference between the average estimated regression coefficient and its true population value did not necessarily decrease as the sample size increased. The average estimated standard deviation of the estimate of the regression coefficient either overestimated or underestimated the actual standard deviation, the former occurring most, but not all, of the time. The significance tests (a two-tailed Z-test with a significance level of 0.05) had actual type I errors ranging from 0.00 to 0.24 for different samples. This approach is therefore inadequate as an epidemiologic tool for analysis of a Framingham-type prospective study. Further simulation studies are indicated.

Coronary Disease↗
PubMed1991

Visuospatial ability, accuracy of size estimation, and bulimic disturbance in a noneating-disordered college sample: a neuropsychological analysis.

The relationship between visuospatial ability and size accuracy in perception was assessed in 69 normal college females. In general, correlations indicated small associations between visuospatial defects and size overestimation and little relationship between visuospatial ability and level of bulimic disturbance. Implications for research on the size overestimation of body image are addressed.

Adolescent↗
PubMed2003

Effect of sample preparation, length of time, and sample size on quantification of total lipids from bovine liver.

The objective was to evaluate the effect of sample preparation (pulverization under liquid nitrogen, homogenization, or sonication), time length of sonication (0-60 s), shaking in chloroform/methanol solvent (0, 2, 4, or 12 h), incubation in chloroform (0 or 12 h), and drying of extracted lipids at 50 degrees C (2, 4, 6, or 24 h), and sample size (50-250 mg) on quantification of total lipids from bovine liver. Pulverization under liquid nitrogen yielded the lowest recovery. Sonication was least time-consuming for sample preparation. Precise estimates and the greatest recovery were obtained with 30 s of sonication, at least 2 h of shaking in chloroform/methanol solvent, 12 h of incubation in chloroform, and at least 6 h of drying. Sample sizes of at least 150 mg gave precise estimates. The results demonstrate that sample preparation, time length of different steps of the extraction procedure, and sample size affect quantification of total lipid from bovine liver.

Animals↗
PubMed1999

A threshold causal model for clinical trials with departures from intended treatment.

Randomized clinical trials often are planned to study a specific intervention. However, the collection of data on treatment actually received often reveals variable levels of treatment exposure (or 'dose') across subjects, due to non-compliance or other reasons. This paper presents a new method, using such 'dose' data as well as control group responses, to assess a causal dose-response relationship. The specific model utilizes a threshold function and incorporates a random effect term to allow for heterogeneous treatment responses among subjects. Further modelling of the random effects allows for reduction of error variance and control for potential confounders. The threshold dose is estimated using a residual variance criterion based on a transformed model. Estimates of standard errors and confidence intervals are obtained using a bootstrap procedure. The method is applied to data from an AIDS clinical trial. A simulation study demonstrates the adequacy of the threshold estimates for particular sample sizes and error variances. The limitations of this essentially exploratory method, as well as some possible extensions, are discussed. Published in 1999 by John Wiley & Sons, Ltd. This article is a US Government Work and is in the public domain in the United States.

CD4 Lymphocyte Count↗
PubMed1982

Experimental design and statistical analysis considerations for in vitro mammalian cell transformation assays with BALB/3T3 cells.

Many mammalian cell assays testing for mutagenic activity have common features which cause statistical estimation and analysis problems. Such assays measure the number of cell alterations occurring in a plate containing an unknown number of cells at risk. The number of cells at risk can be estimated from a parallel cytotoxicity study. While the Poisson distribution has been assumed to apply to standardized frequencies, this is questioned. The failure of standardized frequencies to follow a Poisson distribution is attributed to the relatively small and dosage-dependent number of susceptible cells per plate. A minimum number of such cells per plate or random cluster of plates has been determined for each dose so that the measured variable approximates a Poisson distribution. A transformation is suggested to achieve reasonable normality and variance equality, thereby allowing the use of parametric analysis of variance and regression methods and an estimation of required sample size.

Animals↗
PubMed2006

Interval estimation of the mean response in a log-regression model.

A standard approach to the analysis of skewed response data with concomitant information is to use a log-transformation to normalize the distribution of the response variable and then conduct a log- regression analysis. However, the mean response at original scale is often of interest. El-Shaarawi and Viveros developed an interval estimation of the mean response of a log-regression model based on large sample theory. There is however very little information available in the literature on constructing such estimates when the sample size is small. In this paper, we develop a small-sample corrected interval by using the likelihood-based inference method developed by Barndorff-Nielson and Fraser et al. Simulation results show that the proposed interval provides almost exact coverage probability, even for small samples.

Confidence Intervals↗
PubMed2000

The efficacy of thyroidectomy for Graves' disease: A meta-analysis.

BACKGROUND: Surgery for Graves' disease was largely replaced in the mid-1900s by radioiodine and antithyroid drugs, due to the belief that they were more safe and effective. Since then, thyroid surgery has improved with preoperative drug therapy and modern operative techniques. Recent clinical studies of thyroidectomy for Graves' disease may not reflect outcomes accurately because of small sample size, especially when estimating ideal thyroid remnant size. The purpose of this study was to combine modern clinical trials and use meta-analysis to determine the overall efficacy of both total (TT) and subtotal thyroidectomy (ST) for Graves' disease, compare thyroid function and complications rates of TT and ST, and determine ideal thyroid remnant size. METHODS: Meta-analysis was performed on published studies in which patients underwent either TT or ST for Graves' disease. Meta-analysis was performed by weighted least-squares linear regression. P < 0.05 was considered significant. RESULTS: There were 35 studies comprising 7241 patients. Mean follow-up was 5.6 years. Overall, persistent or recurrent hyperthyroidism occurred in 7.2% of patients. TT was performed on 538 patients and hypothyroidism occurred in all cases. ST was performed in 6703 patients, 59.7% of whom achieved euthyroidism, 25. 6% became hypothyroid, and 7.9% had either persistent or recurrent hyperthyroidism. Permanent recurrent laryngeal nerve injury occurred in 0.9% of TT patients and 0.7% of ST patients (P = NS). Permanent hypoparathyroidism occurred in 1.6% of TT patients and 1.0% of ST patients (P = NS). There was an 8.9% decrease in hypothyroidism and 6.9% increase in euthyroidism for each gram of thyroid remnant (P < 0.0001 each). CONCLUSIONS: Overall, thyroidectomy successfully treated hyperthyroidism in 92% of patients with Graves' disease. There were no cases of hyperthyroidism following TT. ST achieved a euthyroid state in almost 60% of patients with an 8% rate of persistent or recurrent hyperthyroidism. There was no significant difference in complication rates between TT and ST.

Adolescent↗