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 955 records · Page 53Linked to original sources
PubMed1996

Estimated magnitude of behavioral effects of phenytoin in rats and its reproducibility: a collaborative behavioral teratology study in Japan.

A collaborative study was conducted by 30 laboratories that participated in the Behavioral Teratology Meeting in Japan. Pregnant Sprague-Dawley rats from four breeders were orally administrated 200 mg/kg of phenytoin each day from day 10 to day 14 of gestation. The offspring were tested for behavioral teratogenic effects at various ages. The effects were estimated in terms of common effect size, which should be very resistant to the variation inevitable in behavioral teratology results, and thus yield a different type of information from that reported in the usual behavioral teratology studies. The common effect size also gives information on the magnitude of the behavioral teratogenic effects that previous studies could not provide. A breeder difference in the effect of phenytoin for several measures was found in terms of common effect size. The estimated phenytoin effect was found to be large enough to be detected by using a sample size of 20 per group. As to reproducibility of results, estimation by standard deviation across laboratories disclosed that there was almost no difference in magnitude between behavioral and non-behavioral measures, indicating that much of the variation in behavioral teratology results of phenytoin might not be due to measurement error from behavioral tests but rather to the phenytoin effect itself administered via the mother. How seriously a single study is affected by uncontrollable variation of results was illustrated by plotting the respective laboratory results on a dimension. A standard practice in the interpretation of discrepancies among results obtained from single studies is criticized; it is pointed out that such interpretations rely on an optimistic assumption: that is, that results obtained from single studies would be free from the usual variation of results found in behavioral teratogenic research.

Animals↗
PubMed2004

Centile charts II: alternative nonparametric approach for establishing time-specific reference centiles and assessment of the sample size required.

BACKGROUND: Reference intervals, and more generally centile estimates, are used to characterize a reference population for the purposes of interpreting an individual patient's clinical measurement. We describe methods of calculating reference intervals where these centiles vary with a covariate, usually age or time. METHODS: The US Food and Drug Administration and the IFCC have made recommendations on two approaches: the parametric approach, which models the structural characteristics of the data set with a theoretical distribution, and the nonparametric approach, which makes no particular assumption about this structure. In this report we propose a nonparametric procedure that relies on the principles of regression and show how sample size determination can be assessed. We also show how the sample size calculation is influenced by the distribution of the times measured. RESULTS: We illustrated our method on three data sets and compared the results for our proposed nonparametric method with parametric estimates. We showed that the bias is reduced and that the nonparametric method is less likely to produce fluctuating profiles. CONCLUSIONS: To achieve adequate precision the sample size needs to be larger than 120, as has often been recommended. If there is doubt about the parametric model, then threshold sample sizes may need to be as high as 500.

Adolescent↗
PubMed2004

[Isolation of free DNA from archived serum samples suitable for molecular genetic studies].

INTRODUCTION: Collected and archived serum samples could be important sources for genetic studies, once DNA suitable for molecular genetic studies could be obtained from them. METHODS: DNA was isolated from 54 archived sera samples, collected previously from the participants of a Hungarian allergy study, with commercially available isolation kit. The authors have determined the concentration of the isolated DNA (81.88 +/- 52.36 ng/ml) and the size of the isolated fragments was estimated using semiquantitative real-time PCR. Two primers were used producing two different fragment size, for the phospholipase 2A and the actin beta genes, and melting curve analyses was performed as quality control. RESULTS: The concentration of the phospholipase 2A product was 2.9798 +/- 5.4454 microg/microl and the actin beta gene was 0.0015 +/- 0.0011 microg/microl. The melting curve analysis served as a quality control for the determination of the size of PCR products. In the case of the phospholipase 2A all samples produced the 133 bp PCR fragments, except one, while in the case of actin beta gene only six sample showed the expected 178 bp product, all the others samples had smaller fragments. CONCLUSIONS: These results confirm the suitability of the DNA isolated from archived sera samples for further molecular biological studies (SNP analysis, mutation detection) and give an estimate for the product size of the isolated DNA. Sera samples have been collected years ago can be a good source of genetic information on different diseases.

Actins↗
PubMed2003

Comparison of digital photography to weighed and visual estimation of portion sizes.

OBJECTIVE: The primary goal was to test the validity of digital photography for measuring food portion sizes compared with weighed foods and with direct visual estimation. SAMPLES: A total of 60 test meals consisting of 10 different portion sizes from six different university cafeteria menus were prepared and weighed. DESIGN: Food selections and plate waste, as estimated by digital photography and direct visual estimation, were compared with weighed foods. For each method, three observers independently estimated portion sizes of each food. Observers expressed the portion sizes as a percentage of a standard serving. These percentages were multiplied by the weight of the standard portion to yield estimated weights. Statistical analyses To test validity, the estimates of food weights derived from both methods were compared with weighed foods using correlations and were compared with each other using Bland-Altman regression analysis. RESULTS: For the digital photography and direct visual estimation methods, estimates of the portion sizes for food selections, plate waste, and food intake were highly correlated with weighed foods. Both methods tended to yield small overestimates or underestimates. Bland-Altman regression found the two estimation methods to yield comparable results (bias less than 1.5 g). APPLICATIONS/CONCLUSIONS: These findings support the validity of the digital photography method for measuring portion sizes. Digital photography may be most useful for measuring food intake in settings that allow for the direct observation of food selections and plate waste but require minimum disruption of the eating environment, and allow unhurried estimates of portion sizes.

Diet Records↗
PubMed1995

Properties of alternative estimators of familial correlations under variable sibship size.

Selected distributional properties of a variety of estimators of familial correlations (spouse, parent-child, and sibling) were investigated numerically, focusing particularly on sibling correlations under variable sibship size. Maximum likelihood estimators were evaluated for each of the three familial relationships. An additional parent-offspring estimator was studied using a parent-midoffspring estimate pooled over variable sibship size. For sibling intraclass correlations, three estimators based on analysis of variance and three pairwise estimators using different weighting functions for variable sibship sizes were investigated. Correlations were estimated from data simulated under eight sampling conditions (each replicated 1,000 times) using two sets of true parameter values, moderate and large sample sizes, and normal versus highly non-normal sample distributions of data. The estimators are nearly unbiased and efficient, but none of the sibling correlation estimators are normally distributed in small samples. Estimates from highly non-normal data are nearly unbiased but are less efficient than those from normal data.

Bias↗
PubMed1998

Noninvasive methods for quantitating blood time-activity curves from mouse PET images obtained with fluorine-18-fluorodeoxyglucose.

UNLABELLED: The mouse model is currently being explored for various applications with PET imaging. Low resolution of current animal scanners relative to mouse size leads to difficulty in quantitating data from mouse PET images. We have, therefore, investigated methods for determining blood time-activity curves (TACs) from mouse PET studies done with fluorine-18-fluorodeoxyglucose (FDG). METHODS: Eight mice were fasted, the tail vein was injected with 150-300 microCi of FDG and dynamic images were acquired with a CTI/Siemens (Knoxville, TN) animal tomograph for 64.5 min. Concurrently, 11-14 left ventricle (LV) blood samples were drawn directly from the LV chamber. Organ TACs were obtained by drawing circular regions of interest (ROIs) of various sizes on images of the heart, liver and brain. For each mouse, the FDG model parameter K = (K1 x k3)/(k2 + k3) was estimated by a Patlak algorithm with various estimates of the blood TAC and, as a reference tissue TAC, the brain TAC. RESULTS: Most partial-volume-corrected heart ROI TACs overestimated the LV samples. Blood TACs from heart images produced statistically different estimates of K than did the LV samples. The liver image-derived blood TACs yielded estimates of K that were comparable to those yielded by the LV samples. Estimates of K determined with two directly sampled LV points in conjunction with the liver image-derived TAC were not statistically different from the estimates obtained with the LV samples. The size and location of ROIs on images of the liver minimally affected the TACs. CONCLUSION: We have shown that it is experimentally possible to obtain a blood TAC from mouse studies by repeatedly sampling from the LV. We have also shown that images of the liver can be used to reliably estimate the blood TAC. Future FDG PET studies with the mouse model will benefit from this demonstrated ability to noninvasively quantitate blood TACs directly from FDG PET images.

Animals↗
PubMed2003

Data adaptive interim modification of sample sizes for candidate-gene association studies.

OBJECTIVES: The use of conventional Transmission/Disequilibrium tests in the analysis of candidate-gene association studies requires the precise and complete pre-specification of the total number of trios to be sampled to obtain sufficient power at a certain significance level (type I error risk). In most of these studies, very little information about the genetic effect size will be available beforehand and thus it will be difficult to calculate a reasonable sample size. One would therefore wish to reassess the sample size during the course of a study. METHOD: We propose an adaptive group sequential procedure which allows for both early stopping of the study with rejection of the null hypothesis (H0) and for recalculation of the sample size based on interim effect size estimates when H0 cannot be rejected. The applicability of the method which was developed by Müller and Schäfer [Biometrics 2001;57:886-891] in a clinical context is demonstrated by a numerical example. Monte Carlo simulations are performed comparing the adaptive procedure with a fixed sample and a conventional group sequential design. RESULTS: The main advantage of the adaptive procedure is its flexibility to allow for design changes in order to achieve a stabilized power characteristic while controlling the overall type I error and using the information already collected. CONCLUSIONS: Given these advantages, the procedure is a promising alternative to traditional designs.

Data Interpretation, Statistical↗
PubMed2005

On the surveillance for animal diseases in small herds.

Small herds may present a problem in surveillance for infectious animal diseases because typical levels of a within-herd design prevalence are not directly applicable. We suggest a definition of small herds as those smaller than 2/(within-herd design prevalence) on the basis that such herds would be expected to have less than two (i.e. only one) infected animals. Consequently, the probability of detecting small herds cannot be improved by choosing a larger sample size within the herd. We derive necessary sample sizes of herds and the probability ("confidence") of detecting disease within a stratum of small herds, given the among-herd design prevalence and test diagnostic sensitivity. Both a binomial model and a Poisson model can be used to establish the confidence for a given sample size of herds (and vice versa). The results of a simulation study suggest that the Poisson model provides more conservative (lower) estimates of the confidence for a given sample size and should therefore be preferred.

Animal Diseases↗
PubMed1987

Diurnal variation in plasma levels of retinol, tocopherol, and beta-carotene.

Diurnal changes in human plasma levels of the fat-soluble vitamins retinol (R), alpha-tocopherol (E), and beta-carotene (CAR) have not been determined. Plasma levels of these three vitamins in 15 healthy volunteers were measured five times over a 24-hour period. Highly sensitive and specific HPLC assays were used. Mean +/- SEM levels for R, E, and CAR were 591 +/- 36 ng/mL, 10.3 +/- 0.6 micrograms/mL, and 271 +/- 28 ng/mL, respectively. Differences between subjects' mean levels were large and highly significant (p less than 0.0001). Relative to the first 8 AM fasting plasma sample, nonfasting plasma levels drawn over the next 24 hours showed no statistically significant or clinically important changes. The percent increases in standard deviation among subjects due to diurnal variation were 1.5%, 3.3%, and 2.3%, respectively, for R, E, and CAR. This implies that sample sizes for cross-sectional studies investigating differences between two treatments in these vitamin plasma levels would need to be increased by 3%, 7%, and 5%, respectively, if measurements are to be taken throughout the day. Plasma levels may be obtained at random during the day, and need not be 8 AM fasting levels. The small addition in variation introduced by diurnal fluctuations would have minimal impact upon the precision of estimates of treatment effect and sample sizes.

Adult↗
PubMed2006

Procedural justice and layoff survivors' commitment: a quantitative review.

Layoffs are common in today's organizations. Most studies that have examined the correlation between procedural justice and the organizational commitment of layoff survivors have yielded positive correlations, but the magnitude of the correlations varies widely. This study is the first to estimate the population correlation and to identify the primary sources that cause variation in the correlation across studies. The results indicated that justice and commitment correlations can always be expected to be positive. Based on a total sample size of 9080 individuals, the estimated mean population correlation was .34. Variation was primarily explained by attributes of the justice measure where multiple items scales and scales composed of both interactional and procedural justice items yielded higher correlations than single item measures. Therefore, it is important that employers recognize the substantial assuaging affect that procedural and interactional justice can have on survivors' organizational commitment.

Decision Making, Organizational↗
PubMed2002

How much for a star? Elements for a rational choice of sample size in preclinical trials.

In the preclinical field, preliminary estimation of the number of experimental units to be included in a study is an important step in the setting up of the study. This estimation must be based on a dialogue between the experimenter and the statistician. This article, intended for both experimenters and statisticians, is designed to facilitate their dialogue by describing certain elements that are involved in the calculation of the optimal sample size in addition to the factors that influence their quantification. The proposed approach will result in better analysis of each experiment and, therefore, more globally, in rationalization of the use of experimental units.

Drug Evaluation, Preclinical↗
PubMed2002

Two-generation analysis of pollen flow across a landscape. IV. Estimating the dispersal parameter.

The distance of pollen movement is an important determinant of the neighborhood area of plant populations. In earlier studies, we designed a method for estimating the distance of pollen dispersal, on the basis of the analysis of the differentiation among the pollen clouds of a sample of females, spaced across the landscape. The method was based solely on an estimate of the global level of differentiation among the pollen clouds of the total array of sampled females. Here, we develop novel estimators, on the basis of the divergence of pollen clouds for all pairs of females, assuming that an independent estimate of adult population density is available. A simulation study shows that the estimators are all slightly biased, but that most have enough precision to be useful, at least with adequate sample sizes. We show that one of the novel pairwise methods provides estimates that are slightly better than the best global estimate, especially when the markers used have low exclusion probability. The new method can also be generalized to the case where there is no prior information on the density of reproductive adults. In that case, we can jointly estimate the density itself and the pollen dispersal distance, given sufficient sample sizes. The bias of this last estimator is larger and the precision is lower than for those estimates based on independent estimates of density, but the estimate is of some interest, because a meaningful independent estimate of the density of reproducing individuals is difficult to obtain in most cases.

Alleles↗
PubMedDate not supplied

Accuracy and precision of methods to estimate the number of parents contributing to a half-sib progeny array.

Molecular technologies have made feasible large-scale studies of genetic parentage in nature by permitting the genotypic examination of hundreds or thousands of progeny. One common goal of such studies is to estimate the true number of unshared parents who contributed to a large half-sib progeny array. Here we introduce computer programs designed to count the number of gametotypes contributed by unshared parents to each such progeny array, as well as assess the accuracy and precision of various estimators for the true number of unshared parents via computer simulation. These simulations indicate that under most biological conditions (1) a traditional approach (the multilocus MINIMUM METHOD) that merely counts the number of distinct haplotypes in offspring and divides by 2L, where L is the number of loci assayed, often vastly underestimates the true number of unshared parents who contributed to a half-sib progeny array; (2) a recently developed HAPLOTYPES estimator is a considerable improvement over the MINIMUM METHOD when parental numbers are high; and (3) the accuracy and precision of the HAPLOTYPES estimator increase as marker polymorphism and sample size increase, or as reproductive skew and the number of parents contributing to the progeny array decrease. Generally, HAPLOTYPES-based estimates of parental numbers in large half-sib cohorts should improve the characterization of organismal reproductive strategies and mating systems from genetic data.

Animals↗
PubMed1992

Automated DNA profiling by fluorescent labeling of PCR products.

DNA profiling has been automated by the fluorescent tagging of amplified variable number tandem repeat (VNTR) loci. This was achieved by the use of fluorescently labeled primers in the amplification of 10 ng of genomic DNA, coupled with laser detection of the products during electrophoresis. The PCR products are sized by co-electrophoresing a standard size ladder mixed with every sample, thereby eliminating errors in size estimation caused by lane-to-lane differences in migration rate. This increases the precision of VNTR characterization and enables alleles that differ by a single 15-bp repeat to be resolved. The system is capable of high throughput: Twenty-four samples are electrophoresed and analyzed within 6 hr. Also, because four different dyes are available, three different loci can be simultaneously characterized with the fourth dye used for the internal standard. Approximately 100 unrelated British caucasians were analyzed at the loci D1S80, D17S5, and ApoB. The probabilities of two unrelated individuals matching by chance (pM) at these three loci were determined to be 0.065, 0.040, and 0.069, respectively, with a combined pM of 1.8 x 10(-4).

DNA↗
PubMed2003

Bootstrap choice of estimators in parametric and semiparametric families: an extension of EIC.

Ishiguro, Sakamoto, and Kitagawa (1997, Annals of the Institute of Statistical Mathematics 49, 411-434) proposed EIC as an extension of Akaike criterion (AIC); the idea leading to EIC is to correct the bias of the log-likelihood, considered as an estimator of the Kullback-Leibler information, using bootstrap. We develop this criterion for its use in multivariate semiparametric situations, and argue that it can be used for choosing among parametric and semiparametric estimators. A simulation study based on aregression model shows that EIC is better than its competitors although likelihood cross-validation performs nearly as well except for small sample size. Its use is illustrated by estimating the mean evolution of viral RNA levels in a group of infants infected by HIV.

Age Factors↗
PubMed2004

Strategies for two-stage sampling designs for estimating herd-level prevalence.

We propose a herd-level sample-size formula based on a common adjustment for prevalence estimates when diagnostic tests are imperfect. The formula depends on estimates of herd-level sensitivity and specificity. With Monte Carlo simulations, we explored the effects of different intracluster correlations on herd-level sensitivity and specificity. At low prevalence (e.g. 1% of animals infected), herd-level sensitivity increased with increasing intracluster correlation and many herds were classified as positive based only on false-positive test results. Herd-level sensitivity was less affected at higher prevalence (e.g. 20% of animals infected). A real-life example was developed for estimating ovine progressive pneumonia prevalence in sheep. The approach allows researchers to balance the number of herds and the total number of animals sampled by manipulating herd-level test characteristics (such as the number of animals sampled within a herd).

Animals↗
PubMed1999

Distribution and repeatability of anterior pituitary responses to GnRH and relationship of response classification to the postpartum anovulatory interval of beef cows.

Our objectives were to investigate the phenotypic variation in anterior pituitary responsiveness to GnRH (100 microg, i.v.) of beef cows between d 5 and 8 postpartum, estimate repeatability, and determine the relationship between response classification and duration of the postpartum anovulatory interval (PPI). Brahman x Hereford (F1) cows (n = 137) and primiparous heifers (n = 58) were evaluated. Response classifications (Class) included peak LH (Low, Intermediate, or High; Class I) and time to peak LH (Early, 10 to 30 min or Late, 60 to 120 min; Class II). The independent effects of Class I and II on PPI were determined in 145 of 195 cows through twice-weekly serum samples analyzed for progesterone. For Class I, pituitary responses to GnRH approximated a normal distribution and, by definition, differed (P < .001) in magnitudes of peak LH and area under the curve (AUC). For Class II, 111 and 84 cows exhibited early and late peaks, respectively; mean AUC was greater (P < .05) in cows exhibiting late compared with early peaks. Pretreatment LH (P < .01) and estradiol-17beta (P < .004) influenced responses in one or both response classes. Pluriparous cows had shorter (P < .035) PPI than primiparous cows. Class I did not influence the duration of the PPI; however, in Class II, cows with late peaks exhibited an average PPI that was 8 d shorter (P < .025) than in those with an early peak. To estimate repeatability of pituitary responses, 18 classified cows were subsequently rechallenged with GnRH at d 170 of gestation and at the next postpartum period. Although means for each of these challenges differed (P < .05) throughout in both Classes I and II, the small sample size used to make the estimate failed to yield significant (P > .10) interclass correlations. Nevertheless, overall results provide evidence that variability in individual pituitary responses to GnRH could be targeted as a selection marker to improve reproduction.

Animals↗