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 1,243 records · Page 69Linked to original sources
PubMed2001

Modelling the effect of surveillance programmes on spread of bovine herpesvirus 1 between certified cattle herds.

For the eradication of an infectious agent, like bovine herpesvirus 1 (BHV-1), surveillance and certification can be used to reduce the transmission between herds. The goal of surveillance is that a certified herd that becomes infected is detected timely so that infection of several other certified herds is prevented. What counts is whether the reproduction ratio R, i.e. the average number of certified herds infected by one infected certified herd can be kept below 1. To support policy makers in making decisions about the minimal demands for a surveillance programme in an eradication campaign of BHV-1 in cattle, two mathematical models were investigated. With these models, the basic reproduction ratio between herds was calculated. The surveillance programmes were characterised with sample size, sampling frequency, test sensitivity, herd size, vaccination status, and contacts between herds. When R between herds is below 1, then the surveillance programme is sufficiently good to prevent spread of infection, provided that R is estimated well. In the model based on bulk milk testing sample size was replaced by a threshold at which bulk milk can be found positive. The R between herds was mainly influenced by the vaccination status, sampling frequency, and contacts between herds. Herd size moderately affected the outcome. Test sensitivity and sample size, however, were of minor importance. If herds of 50 cows became free of BHV-1 without vaccination, then spread of infection between herds might be prevented when animals within herds are sampled once a year (milk or blood samples). This frequency needs to be intensified, being twice a year, for larger herds and/or herds with extensive contacts with other herds. When bulk milk is sampled instead, sampling should be done at least every 5 months and more intensively, being each month, with larger herd sizes and more contacts between herds.

Animals↗
PubMed2006

Sequential sampling protocols for Spodoptera frugiperda (Lepidoptera: Noctuidae), on Zea mays fields: influence of sampling unit size.

Spodoptera frugiperda (J.E. Smith) is a major pest in maize fields in Argentina. However, a sampling method that accounts for spatial pattern and allows reliable pest density estimations is still lacking. This paper addresses the issue of how sampling unit size can influence the characterization of S. frugiperda spatial pattern and the performance of sampling plans. An intensive sampling programme for S. frugiperda larvae was carried out in maize fields from October until March in four growing seasons. On each sampling date, 12 to 20 sampling units were taken at random. Each unit consisted of 1, 2, 4, 6, 8, 10 or 12 consecutive plants along a row. The linearized version of the Taylor's power law (TPL) was fitted to mean and variance estimated for each sampling date and data set through least squares regression. In general, b values from TPL were significantly >1 (P<0.05), indicating an aggregated sampling distribution. Regression slopes (b) ranged from 1.28 to 1.48 in small larvae and from 1.06 to 1.24 in large larvae, indicating stronger clumping in the former. TPL parameters were used to develop constant precision sampling plans. The efficiency of these plans in terms of expected total cost (searching and handling sampling units) was very sensitive to the size of the sampling units with small larvae but not with large larvae. The influence of sampling unit size and spatial pattern on TPL parameters and sampling costs is discussed.

Animals↗
PubMed1995

Analytical resolution and noise characteristics of linearly reconstructed magnetic resonance data with arbitrary k-space sampling.

The effects of time-varying readout gradients and data sampling with variable dwell times in magnetic resonance imaging are examined. General reconstruction formulas are given for linear reconstruction with even k-space weighting. Closed analytic expressions for estimator variance are given for data sampling during arbitrary gradient waveforms with both uniform kx step size and nonuniform kx step size. It is shown that estimator variance increases (the signal-to-noise ratio decreases) for nonconstant gradient waveforms. It is also shown that estimator variance is greater for constant k-space sampling strategies than for constant time sampling at the Nyquist rate. Data collected during a triangular readout gradient waveform, with either constant time or constant k-space sampling, versus conventional (constant gradient) collection confirms theoretical predictions for estimator variance. The benefits of collecting data while the readout gradient is ramping up from and down to zero are discussed.

Algorithms↗
PubMed2000

Radial variation in sap velocity as a function of stem diameter and sapwood thickness in yellow-poplar trees.

Canopy transpiration and forest water use are frequently estimated as the product of sap velocity and cross-sectional sapwood area. Few studies, however, have considered whether radial variation in sap velocity and the proportion of sapwood active in water transport are significant sources of uncertainty in the extrapolation process. Therefore, radial profiles of sap velocity were examined as a function of stem diameter and sapwood thickness for yellow-poplar (Liriodendron tulipifera L.) trees growing on two adjacent watersheds in eastern Tennessee. The compensation heat pulse velocity technique was used to quantify sap velocity at four equal-area depths in 20 trees that ranged in stem diameter from 15 to 69 cm, and in sapwood thickness from 2.1 to 14.8 cm. Sap velocity was highly dependent on the depth of probe insertion into the sapwood. Rates of sap velocity were greatest for probes located in the two outer sapwood annuli (P1 and P2) and lowest for probes in closest proximity to the heartwood (P3 and P4). Relative sap velocities averaged 0.98 at P1, 0.66 at P2, 0.41 at P3 and 0.35 at P4. Tree-specific sap velocities measured at each of the four probe positions, divided by the maximum sap velocity measured (usually at P1 or P2), indicated that the fraction of sapwood functional in water transport (f(S)) varied between 0.49 and 0.96. There was no relationship between f(S) and sapwood thickness, or between f(S) and stem diameter. The fraction of functional sapwood averaged 0.66 +/- 0.13 for trees on which radial profiles were determined. No significant depth-related differences were observed for sapwood density, which averaged 469 kg m(-3) across all four probe positions. There was, however, a significant decline in sapwood water content between the two outer probe positions (1.04 versus 0.89 kg kg(-1)). This difference was not sufficient to account for the observed radial variation in sap velocity. A Monte-Carlo analysis indicated that the standard error in estimated mean f(S) declined rapidly with increasing sample size. At n = 10, the coefficient of variation in mean f(S) was 7% and at n = 15 it was slightly less than 5%. These observations indicate that radial variation in sap velocity is an important, albeit often overlooked, source of uncertainty in the scaling process. Failure to recognize that not all sapwood is functional in water transport will introduce systematic bias into estimates of both tree and stand water use. Future studies should devise sampling strategies for assessing radial variation in sap velocity and such strategies should be used to identify the magnitude of this variation in a range of non-, diffuse- and ring-porous trees.

Journal Article↗
PubMedDate not supplied

Effects of measurement error on estimating biological half-life.

Direct computation of the observed biological half-life of a toxic compound in a person can lead to an undefined estimate when subsequent concentration measurements are greater than or equal to previous measurements. The likelihood of such an occurrence depends upon the length of time between measurements and the variance (intra-subject biological and inter-sample analytical) associated with the measurements. If the compound is lipophilic the subject's percentage of body fat at the times of measurement can also affect this likelihood. We present formulas for computing a model-predicted half-life estimate and its variance; and we derive expressions for the effect of sample size, measurement error, time between measurements, and any relevant covariates on the variability in model-predicted half-life estimates. We also use statistical modeling to estimate the probability of obtaining an undefined half-life estimate and to compute the expected number of undefined half-life estimates for a sample from a study population. Finally, we illustrate our methods using data from a study of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) exposure among 36 members of Operation Ranch Hand, the Air Force unit responsible for the aerial spraying of Agent Orange in Vietnam.

Environmental Exposure↗
PubMed1996

Reliability and validity of the physical disector method for estimating neuron number.

The physical disector was proposed as an unbiased and efficient means to estimate neuron number; however, the validity and reliability of this method have been examined only infrequently. Estimates of neuron number in the dorsal root ganglia (DRG) of bullfrogs (Rana catesbeiana) were compared to nucleolar counts based on 3-dimensional reconstructions. Accuracy of disector estimates were not affected by size of the animal. Similarly, disector estimates were not systematically altered when area measurements were limited to cellular regions of the DRG versus inclusion of the entire cross-sectional area. However, the recommended protocol for applying the disector resulted in sampling errors that introduced considerable variability in repeated estimates of neuron number from a single ganglion. In addition to this lack of reliability, disector estimates were consistently lower than those obtained by means of a nucleolar counting method that was calibrated against 3-dimensional reconstructions of neuronal profiles. The systematic error of the disector method was greater when ganglia were cut parallel to the long axis of the DR than when they were cut perpendicular to this axis. Increasing the sample size beyond what was recommended increased the reliability of estimates obtained with the disector; however, the bias associated with the plane of section was not reduced. These results emphasize the need for empirical validation of methods used to estimate neuron number in the tissue to which they are to be applied.

Animals↗
PubMed1992

Detection of the relationship between moderate alcoholic beverage consumption and serum levels of estradiol in normal postmenopausal women: effects of alcohol consumption quantitation methods and sample size adequacy.

The purpose of this study was to evaluate the detectability of an effect of moderate alcoholic beverage consumption on the biologic correlate of postmenopausal estradiol levels. Total weekly consumption and beverage-specific intake were assessed using both self-reported usual consumption information and prospective food record data. In terms of total weekly drinks, discrepancies were observed in 35 of the 101 women who reported alcohol use; no consistent pattern of overreporting/underreporting was seen. Although the two alcohol estimates were highly correlated, the relationship between estradiol levels and total weekly alcohol intake was found to be detectable when alcohol consumption based on the food record data was analyzed, but not when the self-report data were examined in a two-tailed hypothesis-testing situation. Evaluating the association between postmenopausal estradiol levels and the two estimates of alcohol intake in random samples of varying sample sizes generated from the mother population of 128 normal postmenopausal women confirmed the finding that the prospectively obtained alcohol data better predict the relationship. Based on the results of this study, it must be concluded that self-reported usual alcohol consumption data must be used with caution when examining an association between alcohol intake and a biologic effect.

Alcohol Drinking↗
PubMed2003

Confidence intervals and sample size calculations for studies of film-reading performance.

The relaxation of restrictions on the type of professions that can report films has resulted in radiographers and other healthcare professionals becoming increasingly involved in image interpretation in areas such as mammography, ultrasound and plain-film radiography. Little attention, however, has been given to sample size determinations concerning film-reading performance characteristics such as sensitivity, specificity and accuracy. Illustrated with hypothetical examples, this paper begins by considering standard errors and confidence intervals for performance characteristics and then discusses methods for determining sample size for studies of film-reading performance. Used appropriately, these approaches should result in studies that produce estimates of film-reading performance with adequate precision and enable investigators to optimize the sample size in their studies for the question they seek to answer.

Clinical Competence↗
PubMed2002

Biomarkers in nutritional epidemiology.

OBJECTIVE: To illustrate biomarkers of diet that can be used to validate estimates of dietary intake in the study of gene-environment interactions in complex diseases. DESIGN: Prospective cohort studies, studies of biomarkers where diet is carefully controlled. SETTING: Free-living individuals, volunteers in metabolic suites. SUBJECTS: Male and female human volunteers. RESULTS: Recent studies using biomarkers have demonstrated substantial differences in the extent of measurement error from those derived by comparison with other methods of dietary assessment. The interaction between nutritional and genetic factors has so far largely gone uninvestigated, but can be studied in epidemiological trials that include collections of biological material. Large sample sizes are required to study interactions, and these are made larger in the presence of measurement errors. CONCLUSIONS: Diet is of key importance in affecting the risk of most chronic diseases in man. Nutritional epidemiology provides the only direct approach to the quantification of risks. The introduction of biomarkers to calibrate the measurement error in dietary reports, and as additional measures of exposure, is a significant development in the effort to improve estimates of the magnitude of the contribution of diet in affecting individual disease risk within populations. The extent of measurement error has important implications for correction for regression dilution and for sample size. The collection of biological samples to improve and validate estimates of exposure, enhance the pursuit of scientific hypotheses, and enable gene-nutrient interactions to be studied, should become the routine in nutritional epidemiology.

Biomarkers↗
PubMed1988

Kernel estimates of dose response.

A nonparametric method for analyzing quantal response data from an indirect bioassay experiment is proposed. Kernel estimates of the dose-response curve are used to develop approximate confidence intervals for (i) the optimal combination dose of a drug with therapeutic effects at low doses and toxic effects at high doses, and (ii) the lethal dose levels of a toxic chemical. This nonparametric procedure was implemented on real and simulated data. The confidence interval for problem (i) has high coverage probabilities when the dose-response curve is symmetric about the optima. However, the coverage probabilities are adversely affected by asymmetry about the optima and consequently are not reliable unless the sample sizes are large. The use of kernel estimators with higher-order kernels may alleviate this sensitivity to asymmetry. The confidence interval for problem (ii) has high coverage probabilities robust with respect to the shape or symmetry of the underlying dose-response curve.

Animals↗
PubMed2006

Some considerations in the design and interpretation of antimalarial drug trials in uncomplicated falciparum malaria.

BACKGROUND: Treatments for uncomplicated falciparum malaria should have high cure rates. The World Health Organization has recently set a target cure rate of 95% assessed at 28 days. The use of more effective drugs, with longer periods of patient follow-up, and parasite genotyping to distinguish recrudescence from reinfection raise issues related to the design and interpretation of antimalarial treatment trials in uncomplicated falciparum malaria which are discussed here. METHODS: The importance of adequate follow-up is presented and the advantages and disadvantages of non-inferiority trials are discussed. The different methods of interpreting trial results are described, and the difficulties created by loss to follow-up and missing or indeterminate genotyping results are reviewed. CONCLUSION: To characterize cure rates adequately assessment of antimalarial drug efficacy in uncomplicated malaria requires a minimum of 28 days and as much as 63 days follow-up after starting treatment. The longer the duration of follow-up in community-based assessments, the greater is the risk that this will be incomplete, and in endemic areas, the greater is the probability of reinfection. Recrudescence can be distinguished from reinfection using PCR genotyping but there are commonly missing or indeterminate results. There is no consensus on how these data should be analysed, and so a variety of approaches have been employed. It is argued that the correct approach to analysing antimalarial drug efficacy assessments is survival analysis, and patients with missing or indeterminate PCR results should either be censored from the analysis, or if there are sufficient data, results should be adjusted based on the identified ratio of new infections to recrudescences at the time of recurrent parasitaemia. Where the estimated cure rates with currently recommended treatments exceed 95%, individual comparisons with new regimens should generally be designed as non-inferiority trials with sample sizes sufficient to determine adequate precision of cure rate estimates (such that the lower 95% confidence interval bound exceeds 90%).

Animals↗
PubMed2001

The impact of nonnormality on full information maximum-likelihood estimation for structural equation models with missing data.

A Monte Carlo simulation examined full information maximum-likelihood estimation (FIML) in structural equation models with nonnormal indicator variables. The impacts of 4 independent variables were examined (missing data algorithm, missing data rate, sample size, and distribution shape) on 4 outcome measures (parameter estimate bias, parameter estimate efficiency, standard error coverage, and model rejection rates). Across missing completely at random and missing at random patterns, FIML parameter estimates involved less bias and were generally more efficient than those of ad hoc missing data techniques. However, similar to complete-data maximum-likelihood estimation in structural equation modeling, standard errors were negatively biased and model rejection rates were inflated. Simulation results suggest that recently developed correctives for missing data (e.g., rescaled statistics and the bootstrap) can mitigate problems that stem from nonnormal data.

Data Collection↗
PubMed2001

Analysis of large and small samples of biochemical and clinical data.

Statistical software often offers a list of various descriptive statistics of location and scale, but rarely selects an efficient estimate that is statistically adequate for an actual univariate sample. The sample interval estimate for a specified degree of uncertainty seems to be more meaningful if it covers an unknown value of the population parameter. The concept of an interval estimate in medicine is then used for medical decision-making. The proposed methodology, which uses the S-Plus algorithm for biochemical, biological and clinical data analysis contains the following steps: (i) Exploratory data analysis identifies basic statistical features and patterns of the data, the distributions of which are mostly non-normal, non-homogeneous and often corrupted by outliers. (ii) Sample assumptions about data, independence of sample elements, normality and homogeneity are examined. (iii) Power transformation and the Box-Cox transformation to improve sample symmetry and stabilize the spread. (iv) Classical and robust statistics for both large (n>30) and medium-sized samples (15<n<30), point and interval estimates for the parameters of location, scale and shape. For an analysis of small samples (4<n<20) the Horn procedure of pivot measures is recommended. The proposed methodology is demonstrated in two case studies, a large sample analysis of mean pregnenolone concentrations in the umbilical blood of newborns, and a small sample analysis of mean haptoglobin concentrations in human serum.

Adult↗
PubMedDate not supplied

An analysis of the geometry of saccular intracranial aneurysms.

BACKGROUND AND PURPOSE: Our goal was to characterize the geometry of simple-lobed cerebral aneurysms and to find the absolute size of these lesions from angiographic tracings. METHODS: Measurements of angiographic neck width (N), dome height (H), dome diameter (D), and semi-axis height (S) were obtained from tracings of 87 simple-lobed lesions located at the basilar bifurcation (BB), middle cerebral (MCA), anterior communicating (AcomA), posterior communicating (PcomA), superior cerebellar (SCA), and posterior cerebral (PCA) arteries. The following ratios were analyzed as subgroups according to location and as a collective sample: dome diameter/dome height (D/H), dome height/neck width (H/N), dome diameter/neck width (D/N), and dome height/semi-axis height (H/S). Using the parent artery as a reference, aneurysm dimensions were normalized to absolute in vivo size. Estimations were validated using angiographic markers. RESULTS: For the entire sample, mean ratios were D/H = 1.11, D/N = 1.91, and H/N = 1.86. For the H/S ratio, the value was 1.98 for BB, MCA, and PcomA lesions and significantly smaller for the AcomA subgroup, at 1.52. The average sizes (in mm) for these dimensions were N = 3.4 for MCA, 3.0 for AcomA, 3.1 for PcomA, and 6.5 for BB; D = 6.1 for MCA, 5.9 for AcomA, 5.3 for PcomA, and 11.7 for BB; H = 5.6 for MCA, 5.0 for AcomA, 5.3 for PcomA, and 11.3 for BB. On average, BB aneurysms were twice as large as aneurysms at other locations. Good correlations were found between the scaled values for D and N, H and N, and H and D. CONCLUSION: These results have been used to characterize the typical simple-lobed aneurysm geometry and to provide a framework for the development of a method of assessment of treatment choice and outcome on the basis of lesion geometry.

Aneurysm, Ruptured↗
PubMed1990

A clinical test for calcium absorption in climacteric women. A two tracer, single blood sample procedure with corrections for body size and skeletal turnover.

This paper reports a two tracers (45Ca for estimating calcium absorption and stable fluoride for assessing skeletal turnover), single blood sample procedure that estimates the absorption of calcium in climacteric women. The proposed technique determines the percentage (Ac) of the dose of 45Ca administered orally (in the fasting state, diluted with 150 ml of milk), present in the extracellular fluid (estimated as 15% of body wt) five hours after intake, corrected for skeletal turnover (assessed by the whole body retention of fluoride). In a series of 13 patients, Ac has been found to be operationally equivalent to the ratio: net Ca absorption/Ca intake, measured under balance conditions. The calcium balances of 60 climacteric women, estimated as: net Ca absorption (Ca intake factored by Ac) minus calciuria was found, as expected, to be correlated with skeletal turnover.

Body Weight↗
PubMed1998

Stability in synapse number and size at 2 hr after long-term potentiation in hippocampal area CA1.

Long-term potentiation (LTP) is an important model for examining synaptic mechanisms of learning and memory. A key question is whether the enhanced synaptic transmission occurring with LTP involves the addition of new synapses, the enlargement of existing synapses, or a redistribution in synaptic weight among synapses. Two experimental designs were used to address this question. In the first experimental design three conditions were evaluated across hippocampal slices maintained in vitro, including slices with LTP analyzed at 2 hr post-tetanus, slices tetanized in the presence of APV, and control slices receiving test stimulation only. In the second experimental design independent LTP and control (low-frequency stimulation) sites were examined. Synapse density was estimated by an unbiased volume sampling procedure. Synapse size was computed by three-dimensional reconstruction from serial electron microscopy (EM). Serial EM also was used to compute synapse number per unit length of dendrite. In both experimental designs there were no significant effects of LTP on total synapse number, on the distribution of different types of synapses (thin, mushroom, stubby, or branched dendritic spines and macular, perforated, or segmented postsynaptic densities), on the frequency of shaft synapses, nor on the relative proportion of single or multiple synapse axonal boutons. There was also no increase in synapse size. These results suggest that LTP does not cause an overall formation of new synapses nor an enlargement of synapses at 2 hr post-tetanus in hippocampal area CA1, and these results support the hypothesis that LTP could involve a redistribution of synaptic weights among existing synapses.

Animals↗
PubMed2003

[Prevalence of symptoms of asthma, allergic rhinitis, conjunctivitis and atopic eczema: ISAAC (International Study of Asthma and Allergies in Childhood) in a population of schoolchildren in Zagreb].

OBJECTIVE: Numerous studies of the population prevalence of asthma, allergic rhinitis, and atopic eczema revealed some international differences. However, the International Study of Asthma and Allergies in Childhood (ISAAC) was the first one using a standardized methodology to evaluate the prevalence of these diseases, and to make comparisons within and between countries. The results showed marked variations in 12-month prevalence of asthma, allergic rhinoconjunctivitis, and atopic eczema symptoms with 20-fold (range 1.6-36.8%), 30-fold (range 1.4-39.7%), and 60-fold (range 0.3-20.5%) differences between the centres with the highest and the lowest prevalence, respectively. AIM: Our aim was to gain the insight into the prevalence of allergic diseases in Zagreb, Croatia by the methods of internationally standardized protocol, proposed by the ISAAC Steering Committee. METHODS: Original questionnaires, translated from English into Croatian, consisting of questions about the child's demographic characteristics, core modules on wheezing, rhinitis and eczema, and supplementary modules, were completed by parents of 10-year-old children (4th grade) attending 18 elementary schools in a city of Zagreb. Total of 1047 questionnaires were returned and analysed after the inconsistent responses were eliminated by phone calling. DISCUSSION: Phase one of the ISAAC study has shown a wide variation in the prevalence of asthma, allergic rhinoconjunctivitis, and atopic eczema symptoms throughout the world, with differing international patterns for the different disorders. Four prevalence ranges have been established for better illustration of the geographic distribution of asthma prevalence: (I) < 5%; (II) 5 to < 10%; (III) 10 to < 20%; (IV) > or = 20%. The highest 12-month prevalences of asthma symptoms were found in developed countries (UK, Australia, New Zealand, Republic of Ireland, and most centres in North, Central, and South America), being in prevalence range IV. The lowest prevalences (range I) were found in several Eastern European countries, Indonesia, Greece, China, Taiwan, Uzbekistan, India, and Ethiopia. According to the results of our study, a continental part of Croatia with a 12-month prevalence of wheezing of 6.02% corresponds to range II. Prevalence of asthma symptoms was greater in males, which is consistent with the results of the younger age group previously analysed. For allergic rhinoconjunctivitis and atopic eczema symptoms grouping of centres with a high prevalences into specific regions was less well defined than for asthma. Centres with the highest prevalences were scattered across the world. In contrast, centres with the lowest prevalences were similar to those for asthma symptoms. Our results of the 12-month prevalence of allergic rhinoconjunctivitis (12.13%), and atopic eczema (7.83%) symptoms were somewhere between the two extremes. As with asthma symptoms, the prevalence of rhinoconjunctivitis symptoms was greater in males. Contrary, the difference in prevalence of atopic eczema symptoms between the sex groups has not been found. The worldwide variations in prevalence of asthma, allergic rhinoconjunctivitis, and atopic eczema symptoms suggest that environmental factors may be critical to the development of these disorders in childhood. Furthermore, different patterns of geographical distribution of particular disorders suggest that major risk factors for them may be different or may involve different latency periods and time trends. Therefore, studies that include objective clinical assessment are required. CONCLUSION: According to our results, Zagreb is a city with relatively low prevalence of allergic diseases symptoms. Larger sample size of at least 3000 subjects is required to provide sufficient precision for estimates of symptom severity, and to generate adequate number of subjects with particular disorders for further analyses. Therefore, we recently increased our sample size to more than 3000 subjects, and started ISAAC Phase two (clinical examination, measures of bronchial hyperresponsiveness, measures of atopy, measures of environmental exposure to aeroallergens, and genetic analyses) in Zagreb, Croatia.

Asthma↗
PubMed2005

Maximum information designs.

BACKGROUND: Expressions to determine the sample size N needed to provide power 1-beta to detect a difference between groups, say delta, involve other nuisance parameters, such as the variance of the observations for a test or means, or the control group probability for a test of proportions or the control group hazard rate for a logrank test of event-times. Designs where N is fixed are called maximum N or duration designs because the sample size and required duration of the study can be fixed, or estimated. However, such designs are expected, but not guaranteed, to provide the desired level of power to detect the specified difference delta at the study end because the true or estimated values of the nuisance parameters are unknown. Thus, the actual information to be accrued and the associated level of power are random variables with sample variation. METHODS: Expressions are developed to determine the amount of information needed to provide the desired level of power, regardless of the values of the nuisance parameters. RESULTS: The amount of information (I) in the observed data is readily quantified and can often be expressed as the inverse of the variance of the test statistic. Also, the total amount of information required to provide the desired level of power 1-beta to detect a difference delta with a test at level alpha is readily determined, designated as Ialpha,beta,delta. Under a maximum information design, the study is continued until the required total amount of information is accrued, or I = Ialpha,beta,delta. In this case, the sample size or duration are random variables, but each can be estimated under various projections about the underlying parameters (variance, probability, hazard) in advance. The implementation of a maximum information design for two and multiple group trials is described for a test of means, proportions and event-times using the logrank test. Application to other methods of analysis is described. CONCLUSIONS: A maximum information design provides greater assurance that the desired level of power will be attained. However, the exact study duration is unknown. Issues related to the implementation of such a design are discussed.

Clinical Trials as Topic↗