Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “STATISTICS”

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
PubMed1985

A model for the statistical description of analytical errors occurring in clinical chemical laboratories with time.

The main purpose of the present study was to describe the statistical behaviour of daily analytical errors in the dimensions of place and time, providing a statistical basis for realistic estimates of the analytical error, and hence allowing the importance of the error and the relative contributions of its different sources to be re-evaluated. The observation material consists of creatinine and glucose results for control sera measured in daily routine quality control in five laboratories for a period of one year. The observation data were processed and computed by means of an automated data processing system. Graphic representations of time series of daily observations, as well as their means and dispersion limits when grouped over various time intervals, were investigated. For partition of the total variation several two-way analyses of variance were done with laboratory and various time classifications as factors. Pooled sets of observations were tested for normality of distribution and for consistency of variances, and the distribution characteristics of error variation in different categories of place and time were compared. Errors were found from the time series to vary typically between days. Due to irregular fluctuations in general and particular seasonal effects in creatinine, stable estimates of means or of dispersions for errors in individual laboratories could not be easily obtained over short periods of time but only from data sets pooled over long intervals (preferably at least one year). Pooled estimates of proportions of intralaboratory variation were relatively low (less than 33%) when the variation was pooled within days. However, when the variation was pooled over longer intervals this proportion increased considerably, even to a maximum of 89-98% (95-98% in each method category) when an outlying laboratory in glucose was omitted, with a concomitant decrease in the interaction component (representing laboratory-dependent variation with time). This indicates that a substantial part of the variation comes from intralaboratory variation with time rather than from constant interlaboratory differences. Normality and consistency of statistical distributions were best achieved in the long-term intralaboratory sets of the data, under which conditions the statistical estimates of error variability were also most characteristic of the individual laboratories rather than necessarily being similar to one another. Mixing of data from different laboratories may give heterogeneous and nonparametric distributions and hence is not advisable.(ABSTRACT TRUNCATED AT 400 WORDS)

Analysis of Variance↗
PubMed1979

[The importance of dietary fibers, also objections against the uncritical use of statistics].

It is explained that mathematical-statistical evalutions in epidemiologic investigations are not principally significant for the causal relationship of different events. This holds true especially for multifactorial connexions. Statistics is an auxiliary science, which is able to point to potential relationships, however, it is not a proof by itself. Unequivocal coincidental relationships can be connected statistically significant. In general the diseases which are related to a lack of fibers in food are typical prosperity diseases (civilization diseases). The same diseases were related earlier to an overconsumption of sugar and they are found also in overnutrition. By selection (e.g. the relation between dietary fibers and coronary heart diseases) and by ignoring important facts the conclusions can be manipulated. It is possible to show in some cases that wrong conclusions are drawn on the basis of a purely mathematical-statistical evalution of nutritional relationships. Even wrong data (e.g. for ethanol consumptions) are evaluated for many years.

Alcoholism↗
PubMed1969

Summary report of a study of academic medical library statistics.

Common statistical analyses were applied to the statistics on medical school libraries published in 1964 and 1966 in the Bulletin to assess the potential utility of such survey data. The major conclusions of the study are that these statistics (1) are highly redundant; (2) are essentially descriptive and not amenable to analysis for predictive purposes; (3) are of questionable reliability; (4) are of minimal utility for library investigators and managers, and of doubtful value for establishing standards. The authors believe that, rather than continuing to collect similar statistics in the future, we need a systematic program to define specific requirements of data use and to design survey methods that will produce data meeting these requirements.

Libraries, Medical↗
PubMed1984

Annual summary of vital statistics--1983.

Data for this article, as in previous reports, are drawn principally from the Monthly Vital Statistics Report, published by the National Center for Health Statistics. The international data come from the Demographic Yearbook and the quarterly Population and Vital Statistics Report, both published by the Statistical Office of the United Nations, which has also been kind enough to provide directly more recent data. Except for mortality data by cause and age, which are based on a 10% sample, all the US data for 1983 are estimates by place of occurrence based upon a count of certificates received in state offices between two dates, one month apart, regardless of when the event occurred. Experience has shown that for the country as a whole the estimates are very close to the subsequent final figures. There are, however, considerable variations in a few of the states, particularly in comparing data by place of occurrence with data by place of residence. State information should be interpreted cautiously.

Adolescent↗
PubMed1984

Recommendations for appropriate statistical practice in toxicologic experiments.

Appropriate statistical practice in toxicologic research is reviewed. The problem centers on the antagonistic needs to discover toxic effects and avoid false indictments of harmless compounds. Specific problems which distort error rates include having many dependent variables, conducting many tests on a variable, data snooping, lack of statistical power and 5) violations of statistical assumptions. Solution strategies include top down planning, designing efficient experiments, balancing type I and type II error rates, selecting hypotheses and 5) using appropriate statistical analysis methods. Top down planning and the "leapfrog" design strategy are particularly emphasized.

Research Design↗
PubMed1984

Retinopathy of prematurity: a risk factor analysis with univariate and multivariate statistics.

Seventy-three out of 639 ophthalmoscopically examined newborns (11.4%) showed varying degrees of retinopathy of prematurity (ROP). Three infants were blind (0.5%). Sixty-six patients were matched with controls according to gestational age, birth weight, birth date and neonatal unit. Using univariate statistics, the variables representing the extent of O2-exposure, elevated paO2, elevated paCO2, paCO2-fluctuations, acidosis, blood transfusions, and artificial ventilation were found to be significantly associated with ROP. In contrast, the variables representing hypocapnia, Hb-levels, parenteral fluid administration and fluid retention showed no correlation with ROP. When a multivariate statistical method was used, the variables representing gestational age, elevated paO2, elevated paCO2 and paCO2-fluctuations as well as the ones defining blood transfusions lost their association with ROP. Episodes of acidosis, multiple birth, birth weight, total hours with Fio2 greater than 0.4, and total duration of artificial ventilation remained in the regression, and hypocapnia gained a significant negative correlation with ROP. This comparative survey based on univariate and multivariate statistics demonstrates that significance levels of identically defined biological and therapeutic variables obtained from the same population of patients are to be interpreted with caution. This fact may also explain contrasting opinions on risk factors in the literature. Multivariate statistical analyses are useful for a relative comparison of significance within a given set of variables. Conclusions on a possible causal relationship as well as on prophylactic measures are not possible.

Humans↗
PubMed1978

Use of a programmable desk-top calculator for the statistical quality control of radioimmunoassays.

We have developed an interactive statistical quality-control system for the small- to medium-sized radioimmunoassay laboratory, which can be used in a programmable desk-top calculator instead of the medium- or large-scale computer systems usually required. The design of this quality-control system is modeled after the suggestions of Rodbard and has three components. The first component evaluates the relationship between the measured response variable of the radioimmunoassay and the precision (or variance) of these measurements. This derived relationship is then used in the second component of the system as the basis for the weighting function used to calculate an interative, weighted, least squares regression of the logit-log transformation of the dose-response curve. The third component uses the quality-control parameters statistically calculated from the linearized dose-response curve to monitor whether the assay is "in-control". The calculator tabulates the means and confidence limits for the various parameters and can plot the statistical quality-control charts. The major benefit of this statistical quality-control system is that it allows the real-time computation and plotting of quality-control data with a programmable desk-top calculator.

Computers↗
PubMed1983

Annual summary of vital statistics--1982.

Data for this article, as in previous reports, are drawn principally from the Monthly Vital Statistics Report, published by the National Center for Health Statistics. The international data come from the Demographic Yearbook and the quarterly Population and Vital Statistics Report, both published by the Statistical Office of the United Nations, which has also been kind enough to provide directly more recent data. Except for mortality data by cause and age, which are based on a 10% sample, all the US data for 1982 are estimates by place of occurrence based upon a count of certificates received in state offices between two dates, one month apart, regardless of when the event occurred. Experience has shown that for the country as a whole the estimates are very close to the subsequent final figures. There are, however, considerable variations in a few of the states, particularly in comparing data by place of occurrence with data by place of residence. State information should be interpreted cautiously.

Birth Rate↗
PubMed1981

Annual summary of vital statistics--1980.

Data for this article, as in previous reports, are drawn principally from the Monthly Vital Statistics Report, published by the National Center for Health Statistics. The international data come from the Demographic Yearbook and the quarterly Population and Vital Statistics Report, both published by the Statistical Office of the United Nations, which has also been kind enough to provide directly more recent data. Except for mortality data by cause and age, which are based on a 10% sample, all the United States data for 1980 are estimates by place of occurrence based upon a count of certificates received in state offices between two dates, one month apart, regardless of when the event occurred. Experience has shown that for the country as a whole the estimate is very close to the subsequent final figures. There are, however, considerable variations in a few of the states, particularly in comparing data by place of occurrence with data by place of residence. State information should be interpreted cautiously.

Adolescent↗
PubMed1980

Annual summary of vital statistics--1979: with some 1930 comparisons.

Data for this article, as in previous reports, are drawn principally from the Monthly Vital Statistics Report, published by the National Center for Health Statistics. The international data come from the Demographic Yearbook and the quarterly Population and Vital Statistics Report, both published by the Statistical Office of the United Nations, which has also been kind enough to provide directly more recent data. Except for mortality data by cause and age, which are based on a 10% sample, all the United States data for 1979 are estimates by place of occurrence based upon a count of certificates received in state offices between two dates, one month apart, regardless of when the event occurred. Experience has shown that for the country as a whole the estimate is very close to the subsequent final figures. There are, however, considerable variations in a few of the states, particularly in comparing data by place of occurrence with data by place of residence. State information should be interpreted cautiously.

Birth Rate↗
PubMed1995

Sequential monitoring of survival data with the Wilcoxon statistic.

When a spending function is used in sequential data monitoring of a clinical trial, it is important to know the information fraction at the times of interim analysis. In a maximum duration designed study, the information fraction is unknown when data are monitored, and it has to be estimated. The modified Wilcoxon statistic developed by Peto and Peto and modified by Prentice is often used to compare two survival curves in a clinical trial. We give guidelines for estimating the information fraction in a maximum duration trial when this statistic is employed. When there is a relatively low event rate or the survival time is approximately exponential, the information fraction for the Peto-Peto-Prentice Wilcoxon statistic is very close to that of the popular logrank statistic. In other cases, it would be helpful to estimate the information fraction as a function of elapsed calendar time. We discuss both group sequential and continuous monitoring.

Biometry↗
PubMed1995

Basic statistics for clinicians: 1. Hypothesis testing.

In the first of a series of four articles the authors explain the statistical concepts of hypothesis testing and p values. In many clinical trials investigators test a null hypothesis that there is no difference between a new treatment and a placebo or between two treatments. The result of a single experiment will almost always show some difference between the experimental and the control groups. Is the difference due to chance, or is it large enough to reject the null hypothesis and conclude that there is a true difference in treatment effects? Statistical tests yield a p value: the probability that the experiment would show a difference as great or greater than that observed if the null hypothesis were true. By convention, p values of less than 0.05 are considered statistically significant, and investigators conclude that there is a real difference. However, the smaller the sample size, the greater the chance of erroneously concluding that the experimental treatment does not differ from the control--in statistical terms, the power of the test may be inadequate. Tests of several outcomes from one set of data may lead to an erroneous conclusion that an outcome is significant if the joint probability of the outcomes is not taken into account. Hypothesis testing has limitations, which will be discussed in the next article in the series.

Clinical Trials as Topic↗
PubMed1994

[Critical comments on the statistical methods in Grawe, Donati and Bernauer: "Change in psychotherapy. From confession to profession"].

In the meta-analysis "Changing Psychotherapy" by Grawe, Donati and Bernauer different psychotherapeutic methods are compared based upon published therapy studies. Hereby the authors claim also to have proven with statistical methods that certain kinds of therapy are more effective than others. I show here that the descriptive and inductive methods used are not able to withstand a critical examination; they are incorrect and in most cases even inadmissible. The results of my examination show that there are four points of critique: 1. The question of how effective a kind of therapy is, according to Grawe's criteria, depends more on the number of variables and their measurements with which a therapy is judged than on the number of patients examined in the single studies. 2. Grawe does not distinguish between dependent and independent variables or measurements; every measurement of each variable is included in his methods with the same weight. 3. The different effect variables used to evaluate the therapic process are mostly represented on varying ordinal scales which are incomparable with each other. Grawe treats these scales as if they were comparable, often even as if they were metric. 4. All five statistical methods (counting significances, binomial test, profile of difference values, t-test, Wilcoxon-test) with which Grawe evaluates the results of the single studies are inadmissible because the conditions required are not met. In sum: The conclusions stated in the meta-analysis cannot be seen as being statistically validated or statistically proven.

Bias↗
PubMed1994

The resampling method of statistical analysis.

Some non-statisticians occasionally use improper statistical methodology due to a lack of appreciation for the model assumptions that underlie a particular technique. The resampling method is a recent attempt to solve statistical problems with a minimum of assumptions. In essence, resampling involves an intuitive approach to inferential statistics that obviates the need for the mathematically derived sampling distribution. The resampling approach takes advantage of readily accessible high-speed computers to do a computationally intensive Monte Carlo experiment on the available data. The resampling approach liberates the user from imposing assumptions that are sometimes dubious. It also directs one away from a black-box attitude toward statistical analysis and instead forces the user to consider the purpose of the inferential process. A particularly user-friendly implementation of resampling methods that addresses some of the problems faced by non-statisticians is found in a simple, yet powerful, computer program called "Resampling Stats," version 3.13.

Adult↗
PubMed1995

[REN plot for a graphic expression method of clinical laboratory data statistics].

Laboratory data have been processed statistically on an assumption that they formed a normal distribution curve by themselves or after their transformation. The data of clinical medicine, however, frequently form distributions different from the normal one and thus they are not exactly represented by simple statistics such as mean and standard deviation. We used non-parametric percentiles(pct) as parameters for an improved expression of data distributions. The parameters used in this method are as follows: mean, 50 pct as median, 25 and 75pct as a concentration indicator, 5 and 95pct as a 90 percent central range, 2.5 and 97.5pct as a 95 percent central range, and minimum and maximum data as a distribution range. These indicators were presented on a new form of box-and-whisker plot. This expression was named as REN plot. Our method was compared to the other statistical expression methods using the data of control surveys of Factor IX activity of normal plasmas and prothrombin time of abnormal standard plasmas as well as APTT of abnormal plasma measured for intra-laboratory quality control. It was found that any form of data distribution, normal or abnormal, was precisely expressed in detail and the gradient of data densities were visually plotted by our method. Also neither the median, central range nor distribution range was influenced by extreme values. It was concluded that REN plot was useful expression for the statistical analysis of clinical laboratory data.

Adult↗
PubMed1994

[A statistical mean dyad test (tt-criteria for comparison of mean trends of two samples)].

The authors claim that widely used statistic tests (Student, Berence-Fisher, Wilcoxon-Mann-Whitney, etc.) may fail to detect differences in average trends. A new statistical tt-test is suggested for such situations when samples are compared belonging to essentially different distributions and one of these distributions is bimodal or U-shaped. This test consists in comparison of the arithmetic means calculated not for the whole sample but separately for values higher and lower than the respective medians. The authors persuasively illustrate the efficacy of this method used in analysis of routine clinico-laboratory data. The statistical mean dyad test (tt-test) is realized in GEMA intellectual system statistical module and used automatically only in cases when routine methods are ineffective.

Data Interpretation, Statistical↗
PubMed1994

[Computer technology of the genogeographic study of the gene pool. I. Statistical information from the genogeographic map].

General statistical information that can be derived from an electronic map and listed in its legend is reviewed. Certain map information (extrema, mean and variance of mapped values, the number of initial and mapped values, their statistical distributions) can be used in analysis not only of gene pool, but also of phene pool maps. Another portion of map information, on total gene diversity, heterozygosity, and interpopulation differentiation, is important for consequent detailed analysis of a gene pool. Statistical information that is derived from a genogeographic map aids understanding of the content of a map and is used in comparative quantitative analysis of different regions and various mapped parameters. The possibilities for interpretation of statistical information of genogeographic maps are shown using the human gene pool as an example. Maps of frequencies of ABO-O and Rh-d blood group genes in three hierarchically subordinate gene pools of native inhabitants of Belarus, the Black Sea-Baltic region, and northeast Eurasia are presented for this purpose.

Computer Simulation↗
PubMed1994

Statistical significance of amino acid sequence similarity in type II DNA methyltransferases.

The statistical significance of amino acid sequence similarities previously observed in type II DNA methyltransferases has been investigated. It is shown: (1) that the intramolecular similarities observed among various type II Mtases are not statistically significant and thus can not be used to support a gene duplication model; (2) that the intermolecular similarities observed in a peptide in various type II adenine methylases are statistically confirmed; (3) that the similarities observed between MutH and these proteins for this peptide are not statistically significant and therefore cannot be used to propose a functional role in DNA recognition for this peptide.

Amino Acid Sequence↗