PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “summary statistics”

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

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

At least 73 records · Page 4Linked to original sources
PubMed1982

Anatomy of the health worker effect - a critique of summary statistics employed in occupational epidemiology.

The "healthy worker effect," perhaps more accurately termed the "active worker effect" has been acknowledged by numerous investigators who have disregarded its significance. The healthy worker effect is expressed by SMRs (standardized mortality ratios) and influenced by the following three factors: (i) selection bias, (ii) improved socioeconomic status, and (iii) the conventional way of calculating SMRs. An examination of these three factors shows that selection for "employability" is probably the most significant factor for the healthy worker effect. For example, the shorter the observation, the smaller the SMR, and the stronger the health worker effect. Secondly, improved socioeconomic status as a result of employment has been shown to lower mortality. Last, although the SMR has the advantage of estimating relative risk in a small sample, it suffers many methodological pitfalls. It is recommended that at least three parameters be used to summarize mortality experience among the employed: (i) relative risk, (ii) attributable risk, and (iii) life expectancy.

Employment↗
PubMed1991

Analyzing laboratory marker changes in AIDS clinical trials.

Repeated measurements of laboratory markers of immunologic or disease status, such as CD4 lymphocyte counts and HIV p24 antigen levels, can be important end points in comparative clinical trials. In this report, we consider comparison of treatment groups with respect to such markers, focusing on a distribution-free approach in which each participant's data are characterized by a single summary statistic. The summary statistics examined are (a) the slope of the least-squares regression of the marker, (b) the average of the last r measurements, and (c) the difference between the averages of the last r and the first s measurements. Under various models of marker time trends, these methods are compared with regard to statistical power. It is found that the slope is usually more efficient than the other two types of summaries. Adaptations for missing data are discussed and illustrated in an analysis of CD4 counts from a recent AIDS clinical trial.

Acquired Immunodeficiency Syndrome↗
PubMed2002

Approximate Bayesian computation in population genetics.

We propose a new method for approximate Bayesian statistical inference on the basis of summary statistics. The method is suited to complex problems that arise in population genetics, extending ideas developed in this setting by earlier authors. Properties of the posterior distribution of a parameter, such as its mean or density curve, are approximated without explicit likelihood calculations. This is achieved by fitting a local-linear regression of simulated parameter values on simulated summary statistics, and then substituting the observed summary statistics into the regression equation. The method combines many of the advantages of Bayesian statistical inference with the computational efficiency of methods based on summary statistics. A key advantage of the method is that the nuisance parameters are automatically integrated out in the simulation step, so that the large numbers of nuisance parameters that arise in population genetics problems can be handled without difficulty. Simulation results indicate computational and statistical efficiency that compares favorably with those of alternative methods previously proposed in the literature. We also compare the relative efficiency of inferences obtained using methods based on summary statistics with those obtained directly from the data using MCMC.

Bayes Theorem↗
PubMed2003

Three methods for analysing correlated ROC curves: a comparison in real data sets from multi-reader, multi-case studies with a factorial design.

This paper compares three published methods for analysing multiple correlated ROC curves: a method using generalized estimating equations with marginal non-proportional ordinal regression models; a method using jackknifed pseudovalues of summary statistics; a method using a corrected F-test from analysis of variance of summary statistics. Use of these methods is illustrated through six real data examples from studies with the common factorial design, that is, multiple readers interpreting images obtained with each test modality on each study subject. The issue of the difference between typical summary statistics and summary statistics from typical ROC curves is explored. The examples also address similarities and differences among the analytical methods. In particular, while point estimates of differences between test modalities are similar, the standard errors of these differences do not agree for all three methods. A simulation study supports the standard errors provided by the generalized estimating equations with marginal non-proportional ordinal regression models.

Analysis of Variance↗
PubMed2003

Relationship between changes in bone mineral density and fracture risk reduction with antiresorptive drugs: some issues with meta-analyses.

UNLABELLED: Published meta-analyses have investigated the relationship between changes in BMD and fracture risk reduction observed with antiresorptive agents, with inconsistent results. Many factors may affect the outcome of such analyses. Our work explores some of these factors and illustrates the need for caution in interpreting the results of meta-analyses. INTRODUCTION: The role of the increase in bone mineral density (BMD) in fracture risk reduction observed in osteoporotic patients treated with antiresorptive drugs is unclear. We examined the effects of study selection, the use of summary statistics or individual patient data (IPD) as the basis for the analyses, and the choice of BMD values used on the outcome of meta-analyses. MATERIALS AND METHODS: To evaluate the effects of study selection, we performed Poisson regression analyses using the results from a number of published studies. To evaluate the effects of using individual patient data instead of summary statistics, we simulated the IPD for vertebral fracture to match the summary statistics for published trials and compared these results with those based on meta-regression using summary statistics. We also evaluated the effect of varying the BMD increase with treatment (3-8%) used in predicting the fracture risk reductions in these simulations. RESULTS: The Poisson regression, which found a statistically significant relationship between nonvertebral fracture risk and spinal BMD when 18 trials of varying designs, duration, and sample size were included in the analysis (p = 0.02), was no longer significant when the analysis was based on the 7 large studies that were placebo-controlled, at least 3 years in duration (at least 1000 patient-years). Meta-analyses of simulated IPD from 12 trials of six antiresorptive agents gave accurate results regardless of the proportion of vertebral risk reduction assumed to be related to BMD change, whereas meta-regression based on summary statistics always produced an estimate around 50%. When the actual data from two risedronate studies were analyzed, the meta-regression based on summary statistics demonstrated a stronger correlation between BMD change and fracture risk reduction than the results based on the IPD analysis. In predicting the fracture risk reduction, the use of the average BMD gain (3%) observed in all studies in the calculations produced an overall fracture risk reduction very similar to the one observed clinically. In contrast, the use of a large BMD gain (8%) produced a substantially higher estimated fracture risk reduction and resulted in a high proportion of fracture risk reduction being attributed to BMD change. CONCLUSIONS: Many factors may influence the outcome of meta-analyses, and caution should be used in interpreting the results of such analyses when exploring the relationship between BMD changes and fracture risk reduction with antiresorptive therapy of osteoporosis.

Bone Density↗
PubMed2003

IS: a web-site for intron statistics.

SUMMARY: A web server has been established for the statistical evaluation of introns in various taxonomic groups and the comparison of taxonomic groups in terms of intron type, length, base composition, etc. The options include the graphic analysis of splice sites and a probability test for exon-shuffling within the selected group. AVAILABILITY: introns.abc.hu, http://www.icgeb.trieste.it/introns

Database Management Systems↗
PubMed1999

Estimating noncancer uncertainty factors: are ratios NOAELs informative?

The prominent role of animal bioassay evidence in environmental regulatory decisions compels a careful characterization of extrapolation uncertainties. In noncancer risk assessment, uncertainty factors are incorporated to account for each of several extrapolations required to convert a bioassay outcome into a putative subthreshold dose for humans. Measures of relative toxicity taken between different dosing regimens, different endpoints, or different species serve as a reference for establishing the uncertainty factors. Ratios of no observed adverse effect levels (NOAELs) have been used for this purpose; statistical summaries of such ratios across sets of chemicals are widely used to guide the setting of uncertainty factors. Given the poor statistical properties of NOAELs, the informativeness of these summary statistics is open to question. To evaluate this, we develop an approach to "calibrate" the ability of NOAEL ratios to reveal true properties of a specified distribution for relative toxicity. A priority of this analysis is to account for dependencies of NOAEL ratios on experimental design and other exogenous factors. Our analysis of NOAEL ratio summary statistics finds (1) that such dependencies are complex and produce pronounced systematic errors and (2) that sampling error associated with typical sample sizes (50 chemicals) is nonnegligible. These uncertainties strongly suggest that NOAEL ratio summary statistics cannot be taken at face value; conclusions based on such ratios reported in well over a dozen published papers should be reconsidered.

Algorithms↗