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 469 records · Page 26Linked to original sources
PubMed2003

Meta-analysis of the effects of endothelin receptor blockade on survival in experimental heart failure.

BACKGROUND: Although an initial study of endothelin receptor blockade reported positive findings, subsequent experiments and clinical trials in humans found little or no benefit. METHODS: We applied meta-analytic methods to assess the methodologic rigor of preclinical studies of endothelin blockade and to quantitatively evaluate the totality of evidence regarding the effect of endothelin receptor blockers in experimental heart failure. A total of 396 animals were assigned to control and 594 were assigned to experimental therapy in the pooled analysis. Of the 9 studies identified, no study reported a priori sample size justification. Although there was a tendency to increased mortality with early administration (relative risk 1.39, P=.15) and decreased mortality with late administration (relative risk 0.85, P=.6), in the overall analysis, there was no significant evidence of benefit or harm (relative risk 1.03, P=.9). Studies with a small sample size had estimated effects that tended to deviate further from the pooled estimate of all studies. CONCLUSIONS: Consideration of mortality effects in the totality of studies revealed no significant effect of endothelin antagonists in animal models of experimental heart failure. Given the potential for between-study variability, reliance on studies with small sample size may lead to unrealistic expectations when extrapolating preclinical experimental results to future research.

Animals↗
PubMedDate not supplied

Sample size recalculation using conditional power.

The sample size required to achieve a given power at a prespecified absolute difference in mean response may depend on one or more nuisance parameters, which are usually unknown. Proposed methods for using an internal pilot to recalculate the sample size using estimates of these parameters have been well studied. Most of these methods ignore the fact that data on the parameter of interest from within this internal pilot will contribute towards the value of the final test statistic. We propose a method which involves recalculating the target sample size by computing the number of further observations required to maintain the probability of rejecting the null hypothesis at the end of the study under the prespecified absolute difference in mean response conditional on the data observed so far. We do this within the framework of a two-group error-spending sequential test, modified so as to prevent inflation of the type I error rate.

Breast Neoplasms↗
PubMed1982

High prevalence of Werner's syndrome in Sardinia. Description of six patients and estimate of the gene frequency.

Several patients with Werner's syndrome in a large family group in Sardinia were ascertained three years ago and reported briefly by Rabbiosi and Borroni (1979). Since then two sisters from a second family and a single case from a third family were ascertained. The three families originated from the Northern part Sardinia and no connection between them was found. We provide a detailed clinical description of six of these patients and attempt to estimate the prevalence and the gene frequency of Werner's syndrome in Sardinia. The prevalence was calculated as 1:94,914 for the two districts of Sassari and Nuoro and as 1:202,766 for the whole island. This is the highest prevalence thus far ascertained. Using Dahlberg's formula we obtained an estimate of the gene frequency q = 0.003288 and thus a frequency of Werner's syndrome of 1:92,515. A more rigorous estimate gave a gene frequency q = 0.001483 and thus a frequency of Werner's syndrome of 1:454,505, but because of the small sample size this estimate should be taken with caution.

Adult↗
PubMed1999

Sample size and power issues in estimating incremental cost-effectiveness ratios from clinical trials data.

It is becoming increasingly more common for a randomized controlled trial of a new therapy to include a prospective economic evaluation. The advantage of such trial-based cost-effectiveness is that conventional principles of statistical inference can be used to quantify uncertainty in the estimate of the incremental cost-effectiveness ratio (ICER). Numerous articles in the recent literature have outlined and compared various approaches for determining confidence intervals for the ICER. In this paper we address the issue of power and sample size in trial-based cost-effectiveness analysis. Our approach is to determine the required sample size to ensure that the resulting confidence interval is narrow enough to distinguish between two regions in the cost-effectiveness plane: one in which the new therapy is considered to be cost-effective and one in which it is not. As a result, for a given sample size, the cost-effectiveness plane is divided into two regions, separated by an ellipse centred at the origin, such that the sample size is adequate only if the truth lies on or outside the ellipse.

Antineoplastic Agents↗
PubMed2005

Two-stage case-control studies: precision of parameter estimates and considerations in selecting sample size.

A two-stage case-control design, in which exposure and outcome are determined for a large sample but covariates are measured on only a subsample, may be much less expensive than a one-stage design of comparable power. However, the methods available to plan the sizes of the stage 1 and stage 2 samples, or to project the precision/power provided by a given configuration, are limited to the case of a binary exposure and a single binary confounder. The authors propose a rearrangement of the components in the variance of the estimator of the log-odds ratio. This formulation makes it possible to plan sample sizes/precision by including variance inflation factors to deal with several confounding factors. A practical variance bound is derived for two-stage case-control studies, where confounding variables are binary, while an empirical investigation is used to anticipate the additional sample size requirements when these variables are quantitative. Two methods are suggested for sample size planning based on a quantitative, rather than binary, exposure.

Case-Control Studies↗
PubMed2003

Within-plant distribution of twospotted spider mite, Tetranychus urticae Koch (Acari: Tetranychidae), on ivy geranium: development of a presence-absence sampling plan.

The twospotted spider mite, Tetranychus urticae Koch, is an important pest of ivy geranium and other ornamental plants. As a part of our long-term goal to develop an integrated crop management program for ivy geraniums, the focus of this study was to produce a reliable sampling method for T. urticae on this bedding plant. Within-plant mite distribution data from a greenhouse experiment were used to identify the young-fully-opened leaf as the sampling unit. We found that 53% of the mites on a plant are on the young-fully-opened leaves. On average 22, 37, and 41% of the leaves belonged to the young, young-fully-opened, and old leaf categories, respectively. We then developed a presence-absence sampling method for T. urticae in ivy geranium using generic Taylor's coefficients for this pest. We found the optimal binomial sample sizes for estimating populations of T. urticae at densities of between 0 and 3 mites/leaf to be quite large; therefore, we recommend the use of numerical sampling within this range of T. urticae densities. We also suggest that population estimates of T. urticae on ivy geranium be done based on mite density/unit area of greenhouse space, both for conventional greenhouse pest management, and for determining how many phytoseid predators to release when using biological control.

Animals↗
PubMed2006

A simple method for assessing sample sizes in microarray experiments.

BACKGROUND: In this short article, we discuss a simple method for assessing sample size requirements in microarray experiments. RESULTS: Our method starts with the output from a permutation-based analysis for a set of pilot data, e.g. from the SAM package. Then for a given hypothesized mean difference and various samples sizes, we estimate the false discovery rate and false negative rate of a list of genes; these are also interpretable as per gene power and type I error. We also discuss application of our method to other kinds of response variables, for example survival outcomes. CONCLUSION: Our method seems to be useful for sample size assessment in microarray experiments.

Computer Simulation↗
PubMed1999

Empiric assessment of parameters that affect the design of multireader receiver operating characteristic studies.

RATIONALE AND OBJECTIVES: The authors attempted to assess experimentally the magnitude of reader variability and the correlations and interactions among cases, readers, and modalities during observer performance studies and their possible effects on study design and sample size. MATERIALS AND METHODS: Published data from 32 selected receiver operating characteristic (ROC) studies were reviewed to compare the magnitude of the variance component from readers with the variance component from modality. Estimates of correlation and interactions among cases, readers, and modalities were also computed directly from ROC data ascertained during two large studies performed in our laboratory. Each of these two studies included 529 cases and six readers, but one study used eight modalities and the other nine. RESULTS: Published results indicate that reader variability is task dependent and larger (P < .05) than modality variability in detection of interstitial disease. Measured correlations between modalities for the same reader were task dependent and ranged from 0.35 to 0.59. Modality-by-reader and modality-by-case interactions often are not important factors. The random error term was greater than the modality-by-reader interaction in 11 of 20 comparisons and greater than the modality-by-case interaction in eight of 20 comparisons. CONCLUSION: Use of the same cases interpreted with different modes is justifiable in many situations because of the high variability from readers. This comprehensive review of existing ROC studies resulted in parameter assessments that can be used to better estimate sample-size requirements in multireader ROC studies.

Humans↗
PubMed2002

The variability of estimates of variance, and its effect on power analysis in monitoring design.

Power analysis can be a valuable aid in the design of monitoring programs. It requires an estimate of variance, which may come from a pilot study or an existing study in a similar habitat. For marine benthic infauna, natural variation in abundances can be considerable, raising the question of reliability of variance estimates. We used two existing monitoring programs to generate multiple estimates of variance. These estimates were found to differ from nominated best estimates by 50% or more in 43% of cases, in turn leading to under or over-estimation of sample size in the design of a notional monitoring program. The two studies, from the same general area, using the same sampling methods and spanning a similar time scale, gave estimates varying by more than an order of magnitude for 25% of taxa. We suggest that pilot studies for ecological monitoring programs of marine infauna should include at least two sampling times.

Ecology↗
PubMed2004

A meta-analysis of the effect of mediated health communication campaigns on behavior change in the United States.

A meta-analysis was performed of studies of mediated health campaigns in the United States in order to examine the effects of the campaigns on behavior change. Mediated health campaigns have small measurable effects in the short-term. Campaign effect sizes varied by the type of behavior: r=.15 for seat belt use, r=.13 for oral health, r=.09 for alcohol use reduction, r=.05 for heart disease prevention, r=.05 for smoking, r=.04 for mammography and cervical cancer screening, and r=.04 for sexual behaviors. Campaigns with an enforcement component were more effective than those without. To predict campaign effect sizes for topics other than those listed above, researchers can take into account whether the behavior in a cessation campaign was addictive, and whether the campaign promoted the commencement of a new behavior, versus cessation of an old behavior, or prevention of a new undesirable behavior. Given the small campaign effect sizes, campaign planners should set modest goals for future campaigns. The results can also be useful to evaluators as a benchmark for campaign effects and to help estimate necessary sample size.

Communication↗
PubMed1995

Calibration of dietary intake measurements in prospective cohort studies.

To evaluate the accuracy of dietary intake measurements in prospective cohort studies on diet, it is generally proposed that substudies be conducted to 1) correct relative risk estimates for biases due to measurement error, and 2) account for statistical power losses when estimating the sample size requirements of the cohort. Usually the substudy takes the form of a "validity" study, based on a small group of volunteers and using repeated daily food consumption records as reference measurements. In this methodological review, the authors conclude that when relative risks are estimated for scaled, absolute intake differences rather than for quantile categories, a "calibration" study based on only a single day's food intake record (but generally on a larger number of subjects) can provide sufficient reference information to meet objectives 1 and 2. A major advantage of calibration studies based on this single-day-per-subject design is that they can be conducted on a representative sample of cohort participants more easily than validity studies in which reference measurements are repeated.

Bias↗
PubMed1991

Confidence intervals and sample sizes.

In a recent paper, Beal (1989, Biometrics 45, 969-977) considers the problem of determining the appropriate sample size when inference about a parameter theta is to be made on the basis of a confidence interval (CI). He suggests that the sample size should be chosen so that the probability that the length of the CI is less than a given value, conditional on the interval including the true theta, is greater than a specified level. In this note, in which we concentrate on two-sided intervals, this suggestion is examined, as is the effect of uncertainty in our knowledge of the population variance sigma 2 on estimates of sample size.

Biometry↗
PubMed1978

An application of multivariate ratio methods for the analysis of a longitudinal clinical trial with missing data.

This paper presents an analysis of a longitudinal multi-center clinical trial with missing data. It illustrates the application, the appropriateness, and the limitations of a straightforward ratio estimation procedure for dealing with multivariate situations in which missing data occur at random and with small probability. The parameter estimates are computed via matrix operators such as those used for the generalized least squares analysis of catetorical data. Thus, the estimates may be conveniently analyzed by asymptotic regression methods within the same computer program which computes the estimates, provided that the sample size is sufficiently computer program which computes the estimates, provided that the sample size is sufficiently large.

Clinical Trials as Topic↗
PubMed2005

Feasibility of a randomized trial of extended lymphadenectomy for pancreatic cancer.

HYPOTHESIS: The required sample size of a prospective randomized trial comparing standard pancreaticoduodenectomy with pancreaticoduodenectomy plus extended lymphadenectomy for pancreatic adenocarcinoma is prohibitively large, making such a trial infeasible. DESIGN: Retrospective cohort study. SETTING: Comprehensive cancer center. PATIENTS: We identified 158 patients who underwent pancreaticoduodenectomy for pancreatic adenocarcinoma with separate pathologic analysis of second-echelon lymph nodes, defined as lymph nodes along the proximal hepatic artery and/or the great vessels. MAIN OUTCOME MEASURES: To estimate the sample size required for a randomized trial, we devised a biostatistical model with the following assumptions: extended lymphadenectomy can benefit only patients who (1) actually have disease removed from second-echelon nodes, (2) have microscopically negative (R0) primary tumor resection margins, and (3) do not have visceral metastatic (M0) disease. RESULTS: Seventy-six patients (48.1%) had negative first- and second-echelon lymph nodes, 65 (41.1%) had positive first-echelon and negative second-echelon lymph nodes, and 17 (10.8%) had positive first- and second-echelon lymph nodes. Patients with positive second-echelon lymph nodes had an R0 resection rate of 47.1%. At a median follow-up of 65.1 months, 4 patients with positive second-echelon lymph nodes were alive, but 3 had recurrent disease. This implies that only 1 patient (5.9%) with positive second-echelon lymph nodes may have had true M0 disease. Therefore, only 0.3% of patients (10.8% with positive second-echelon lymph nodes x 47.1% with R0 resection x 5.9% with M0 disease) may achieve a survival benefit from extended lymphadenectomy. A randomized trial of standard pancreaticoduodenectomy vs pancreaticoduodenectomy with extended lymphadenectomy would require 202 000 patients in each study arm to detect such a small difference. CONCLUSIONS: Definitive evaluation of the potential benefits of extended lymphadenectomy would require a prohibitively large sample size. Adequately powered randomized trials to address the potential benefit of extended lymphadenectomy are infeasible.

Adenocarcinoma↗
PubMed2004

Sample size calculations for controlled clinical trials using generalized estimating equations (GEE).

OBJECTIVES: Clinical trials with correlated response data based on generalized estimating equations (GEE) have become increasingly popular as they require smaller samples than classical methods that ignore the clustered nature of the data. We have recently derived the recommendation to use the independence estimating equations (IEE) as primary analysis in most controlled clinical trials instead of GEE with estimated correlations. Although several approaches for sample size and power calculation have been proposed, we have shown that most of these procedures are very specific and not as general as required for designing clinical trials. METHODS: We extended the previously developed SAS macro GEESIZE to overcome this restriction. Specifically, we have added the option of an independence working correlation matrix required for the IEE. Additionally, we have reformulated the hypotheses to allow for coding that includes an intercept term instead of the previously used analysis of variance coding. RESULTS: To demonstrate the validity of GEESIZE we investigate the calculated sample sizes for specific models where closed formulae are available. For illustration, we utilize GEESIZE for planning a new trial on the treatment of hypertension and thereby exemplify its flexibility. CONCLUSIONS: We show that our freely available macro is a very general and useful tool for sample size calculation purposes in clinical trials with correlated data.

Cluster Analysis↗
PubMed2002

Health-related consequences of overactive bladder.

OBJECTIVE: Overactive bladder (OAB) is a condition of urgency, with or without urge incontinence, usually with frequency and nocturia. This study assesses whether people with OAB are at greater risk for urinary tract infections (UTIs), falls and injuries, and increased number of visits to the doctor compared to age- and gender-matched controls. The study also estimates costs associated with these health-related consequences. PATIENTS & METHODS: A US representative telephone survey under the National Overactive Bladder Evaluation (NOBLE) Program was conducted with 5204 English-speaking adults older than 18 years. The survey asked respondents about bladder symptoms. Based on the telephone survey, 865 symptom-identified OAB cases and 903 age- and gender-matched controls were sent a postal questionnaire. A total of 397 cases and 522 controls returned the questionnaires. Nonrespondent cases and controls did not differ with regard to age, gender, educational status, diabetes, congestive heart failure, and self-rated health status. Regression analyses were conducted to assess the effect of OAB on health-related consequences, controlling for age, gender, race, education, marital status, number of previous births, self-reported health status, diabetes, and congestive heart failure. RESULTS: People with OAB reported 0.84 (20%) more visits to the physician (P < .05) and 0.21 (138%) more UTIs in the last year than people without OAB (P < .001). Overactive bladder cases also had over twice the odds of being injured in a fall than people without OAB (odds ratio = 2.26; 95% confidence interval 1.46, 3.51). Consistent with having more falls, OAB cases had an increased risk of bone fracture (P < .1). This effect, however, was not statistically significant (at alpha level 0.05) due to the limited sample size. The estimated cost of UTIs associated with OAB was approximately $1.37 billion US dollars in year 2000. The cost of falls without bone fracture due to OAB was $55 million. Falls with bone fracture accounted for approximately $386 million; however, further research with a larger sample is needed to accurately estimate these costs. CONCLUSION: People with OAB self-report significantly more UTIs and a greater risk of being injured in a fall. Given the large prevalence of UTIs and concerns of overprescribing antibiotics, these results are important for health plans and policy makers. In addition, people with OAB visit their physicians more often than people without OAB. These consequences entail significant economic costs, of which a large percentage will be incurred by health plans. To the extent that OAB causes these consequences, there may be significant savings from effectively treating OAB.

Accidental Falls↗
PubMed1993

Measuring intrasubject variability: use of the jacknife in doubly labeled water experiments.

The doubly labeled water technique measures energy expenditure; however, very little has appeared in the literature regarding estimation of the intrasubject variation. By use of a statistical resampling procedure called the jackknife, the standard deviation of the determination of energy expenditure in each subject is evaluated. Jackknife methods exploit the regression techniques that are already used with the doubly labeled water technique and are very easy to implement. Estimates of sample sizes for future experiments can easily be done with the jackknife. These formulas give the number of determinations of isotopic enrichment of hydrogen and oxygen over time that are needed to achieve a given degree of accuracy in estimating energy expenditure. An example with two human subjects illustrates the methodology of the jackknife.

Body Water↗
PubMed2004

Effect of aggregation of horn fly populations within cattle herds and consequences for sampling to obtain unbiased estimates of abundance.

Reanalysis of counts of horn fly, Hematobia irritans (L.), obtained from a variety of cattle herds indicated that aggregation of the flies within herds decreased as mean fly density increased. Aggregation was also related to the proportion of fly-resistant and fly-susceptible cattle in a herd. Herds were grouped according to their degree of horn fly aggregation. Low aggregation herds included larger framed Angus, Horned Hereford, Polled Hereford, and Red Poll breeds. Moderate aggregation occurred with Brahman, Charolais, small-framed Angus, mixed cows, and Hereford x Charolais cross. High aggregation occurred with Chianina and mixed herds. Relationships between the sample means and variances varied among aggregation groups. A resampling approach was used to determine the influence of random sampling of a herd on the proportion of horn fly population estimates within fixed percentages of the true mean. The proportion of sample means within +/- 5, 10, 15, and 20% of the true means varied with the proportion of the herd sampled, the mean and variance of fly density, and herd size. Recommendations for obtaining sample size to estimate fly density within a fixed percentage of the true mean are given.

Alberta↗