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 523 records · Page 29Linked to original sources
PubMed1975

[Plea for the application of statistical methods by computer to stomatological research].

In face of the double reflex of fascination and defiance which is right away created by statistical methods by computers, the authors have attempted by means of precise examples to demystify these methods by recalling: 1) That computer statistics can be defined as the association of a service and a machine, the combination should be considered as an instrument and only an instrument. 2) That the application of computer statistics to odonto-stomatological research of course evidently enables the resolution of problems which would be otherwise insoluble, but above all considerably enlarges the field of research by rendering visible phenomena which would otherwise remain hidden. In this way, statistical methods by computer can be compared with the microscope; a low magnification with the optical microscope justifies only an overall view of the section while changing to higher magnifications permits the demonstration of details which would remain unsuspected. Use of the electromicroscope opens up another world for research. 3) In the same way as the biologist who uses the microscope, if he knows the principle of it is not in so far a specialist in optics, neither does the research woker who uses statistics necessarily have to be a trained statistician, on the one condition that evidently he disposes of professional computer logistics. This is the case of the ACB (Nantes) rush technical units, who cosigned this work, where the engineers who pose the problems to be resolved and emit the working hypotheses from the results supplied by the statistical studies are not the same people who carry out these very studies. 4) An essential fact, the research worker's independance is complete, only he can pose the problems and emit the hypothesis from the correlation, i. e. point of fact concomitance between the various isolated parameters. It therefore does not seem that weighty statistical techniques presently occupy the place of choice which should be theirs in stomatological research and more particularly in research concerning growth and development of the face where their use seems to be electively implied in face of the number and complexity of the parameters looked at.

Computers↗
PubMed2000

A structured curriculum for improved resident education in statistics.

Many resident physicians do not receive sufficient education in statistical theory and study design to allow them to effectively interpret and apply the medical literature to patient care. A survey of 62 surgical residency programs identified that only 33 per cent include formal statistics teaching in their curricula. A structured curriculum was formulated to introduce surgical residents to basic statistical theory, common statistical tests, and study design. Lectures were integrated into an existing monthly journal club in which manuscripts reviewed were used to illustrate the statistical concepts taught. Knowledge improvement was determined using a multiple choice test given before and after completion of the curriculum. Mean test scores increased significantly as a result of the curriculum (P < 0.004). Formal statistics education using a structured curriculum significantly improves resident physician knowledge of statistical theory and study design. Such teaching can be easily integrated into a residency program's existing curriculum.

Curriculum↗
PubMed2004

[A review of the statistical methods used in the journal of prosthetic dentistry].

PURPOSE: To understand the present condition of statistical methods used in the Journal of Prosthetic Dentistry. METHODS: The statistical methods used in 448 treatises (from 2001 to 2003) in the journals were reviewed, classified and analysed. RESULTS: The utilization rates of statistical methods were 28.06%, 48.34% and 51.02% from 2001 to 2003,respectively. The utilization rate of statistical methods was increasing significantly with year (chi2=29.453,P<0.001). The main statistical methods were analysis of variance (ANOVA), multiple comparison, nonparametric test, t-test, linear regression and correlation and chi2 test. CONCLUSION: The utilization rates of sophisticated statistical methods and the proportion of using more than two kinds of statistical tests were all higher than that in the medical journals published in China.

Humans↗
PubMedDate not supplied

[Statistical indices in medicine and epidemiology and their classification].

Mathematical statistics deals with abstract notions, while medicine solves complicated and many-sided problems. For this reason medical statistics faces some moot points in the interpretation of a number of notions and the classification of statistical indices. In the present article the definition of variables and statistical indices is formulated and their characterization is given. An attempt is made to provide the systematization and natural classification of the latter. Statistical indices are defined as the characteristics of statistical totalities. To classify statistical indices, the most essential signs are used: the character of a variable (external relations), the trend of study (internal content), the form of expression (calculation), derived indices and characteristics (comparison and the results of analysis).

Algorithms↗
PubMed1988

A new statistical test for linkage heterogeneity.

A new, statistical test for linkage heterogeneity is described. It is a likelihood-ratio test based on a beta distribution for the prior distribution of the recombination fraction among families (or individuals). The null distribution for this statistic (called the B-test) is derived under a broad range of circumstances. Two other heterogeneity test statistics--the admixture test or A-test first described by Smith and Morton's test (here referred to as the K-test)--are also examined. The probability distribution for the K-test statistic is very sensitive to family size, whereas the other two statistics are not. All three statistics are somewhat sensitive to the magnitude of the recombination fraction theta. Critical values for each of the test statistics are given. A conservative approximation for both the A-test and B-test is given by a chi 2 distribution when P/2 instead of P is used for the observed significance level. In terms of power, the B-test performs best among the three tests over a broad range of alternate heterogeneity hypotheses--except for the specific case of admixture with loose linkage, in which the A-test performs best. Overall, the difference in power among the three tests is not large. An application to some recently published data on the fragile-X syndrome and X-chromosome markers is given.

Chromosome Fragility↗
PubMed1987

A comparison of hospital admissions data and official government statistics of serious traffic accident injuries.

Two sources of New Zealand statistics on serious traffic accident injuries were compared for the period 1973-82. The official statistics based on Ministry of Transport reports showed consistently lower figures than the supposedly comparable hospital admission statistics from the Department of Health. The reporting ratio in the official statistics had decreased from 66% of the first admissions to hospital in 1973 to 43% in 1982. The highest reporting ratios were observed for drivers and passengers (66% in 1982). For pedestrians it was 48%, for motor cyclists 45% and for bicyclists 15%. A detailed analysis of reporting ratios for specific types of accidents among children showed that for certain accidents almost none of the injured children were recorded in the Ministry of Transport statistics. It is concluded that these statistics present a distorted picture of the nonfatal road toll and that a new statistical reporting system combining information from several sources is needed.

Accidents, Traffic↗
PubMed1983

Statistical procedures in clinical pharmacology: decision-theorectic foundations.

The function of statistical analysis is to reach optimal decisions in situations of principal uncertainty. The foundations of statistical procedures are experiments X, whose distributions PX are characterised by random samples. Within given classes of distributions (distribution hypothesis) mathematical statistics make optimal procedures available, which allow conclusions on the real properties of PX. The use of optimal procedures minimises the risk of incorrect decisions made on the basis of random samples (principle of greatest chance). Statistics, however, refrain from any statement on the quality of the decision to which a procedure in a certain situation leads. Against a decision-theoretic background of statistical decisions the growing complexity of clinical-pharmacological test models leads to the necessity of using descriptive-statistical methods on an even larger scale. The aim should be to reduce the number of individual parameters to a few important factors, so that the relation between the number of trial subjects on the one hand and the number of parameters on the other remains valid for meaningful statistical statements.

Clinical Trials as Topic↗
PubMed1993

The statistical power of family practice research.

To determine the statistical power of family practice research, we conducted a survey of all family practice studies published in 3 major family practice journals in 1988. Only 5 of 86 studies calculated statistical power, although 67 (78%) reported statistically nonsignificant results. Calculations of statistical power revealed that more than 80% of family practice studies had sufficient statistical power to detect medium to large effect sizes, but not small effect sizes, which some readers might deem clinically important. We conclude that the authors of family practice research studies do not pay enough attention to the issue of statistical power when reporting statistically nonsignificant results, leaving their readers in doubt as to whether a clinically significant effect could have been detected.

Data Interpretation, Statistical↗
PubMedDate not supplied

[Statistics and medical decision making].

Statistics is the science that studies variability in all its forms and, in essence, is an instrument that allows us to analyze variable phenomena and detect small, but nevertheless important, differences. The instrumental character of this science made Statistics a fundamental tool for the study of biological and clinical phenomena, but the purpose of this discussion is to focus on the fact that Statistics, either directly or indirectly, is presently the basis of virtually all medical decision-making processes and, consequently, has had a major influence on clinical practice itself. Disease prevalence, the consideration of risk factors, for instance, which is information of great importance in the process of diagnosis, is obtained through studies that are based in statistical methodologies. It is, however, in the decision-intensive processes of selecting diagnostic examinations, choosing the best therapy and evaluating the prognostic implications, that Statistics have had a more direct impact on clinical practice. In modern Medicine, Statistics may no longer be seen as a set of techniques that are applied in scientific research. Rather, Statistics should be understood as the appropriate language for describing and discussing observed phenomena.

Diagnosis↗
PubMed1998

Identifying measurement disturbance effects using Rasch item fit statistics and the Logit Residual Index.

A Monte Carlo study was conducted using simulated dichotomous data to determine the effects of guessing on Rasch item fit statistics (weighted total, unweighted total, and unweighted between fit statistics) and the Logit Residual Index (LRI). The data were simulated using 100 items, 100 persons, three levels of guessing (0%, 25%, and 50%), and two item difficulty distributions (normal and uniform). The results of the study indicated that no significant differences were found between the mean Rasch item fit statistics for each distribution type as the probability of guessing the correct answer increased. The mean item scores differed significantly with uniformly distributed item difficulties, but not normally distributed item difficulties. The LRI was more sensitive to large positive item misfit values associated with the unweighted total fit statistic than to similar values associated with the weighted total fit or unweighted between fit statistics. The greatest magnitude of change in LRI values (negative) was observed when the unweighted total fit statistic had large positive values greater than 2.4. The LRI statistic was most useful in identifying the linear trend in the residuals for each item, thereby indicating differences in ability groups, i.e. differential item functioning.

Bias↗
PubMed2002

[Statistics must be seen and scrutinized! Fundamental to the progress of medical science--yet neglected in medical journals].

Even the most respectable medical journals contain questionable conclusions, frequently based on erroneous statistical analyses. A common attitude among medical researchers seems to be that a deeper knowledge of statistical theory is not necessary--it might suffice to be acquainted with the statistical program packages. The most common statistical errors in medical manuscripts are related either to the structure of the data or to misunderstandings regarding the logic in some basic statistical procedures, especially significance analysis. Editors should designate professional statisticians to sort out those manuscripts which require further scrutiny from a statistical point of view. This can be organized on a consultative basis outside the framework of the traditional, "blinded" peer-review system. The assessments made by the medical journals have come to play a very specific role in the Swedish academic system with respect to doctoral dissertations. Further discussion of present procedures in this area would be highly desirable.

Data Interpretation, Statistical↗
PubMed2002

Mantel statistics to correlate gene expression levels from microarrays with clinical covariates.

Mantel statistics provide an additional step to standard approaches in the analysis of gene expression and covariate data, allow the calculation of standard statistics such as correlation, partial correlation, and regression coefficients, and, with permutation tests, provide P values for these statistics to relate the sample covariates to the expression levels. In this article we describe the Mantel statistics and illustrate their use and interpretation with data from a study of seven human oligodendrogliomas (brain tumors) where expression levels of 1013 genes and five covariates were previously analyzed using standard approaches. In the previous analysis of these data, qualitative relationships were found between gene expressions and two of the clinical covariates. We show in this article how the Mantel statistics are able to formally quantify and provide P values to determine statistical significance of these relationships. We also show how the Mantel statistics can be used to rank subsets of genes, found using standard clustering methods, in terms of differential expression across samples.

Adult↗
PubMed2001

Two approaches for consolidating results from genome scans of complex traits: selection methods and scan statistics.

This work has two purposes: (i) empirically selecting levels of significance that maximize the fraction of markers close to a gene (hit rate) when performing linkage analyses of simulated data and (ii) evaluating the utility of a previously reported scan statistic on the same data. Genotype data were simulated from a trait model of seven susceptibility genes. For purpose (i), five statistics were evaluated on all marker loci in fifty replicates; two-point lod and heterogeneity lod scores maximized over dominance (mlod, mhlod), a multi-allelic TDT test, an affected sib-pair test (ASP), and a model-free test on all sib-pairs (ALL_SIBS). Within each replicate the fraction of markers (hit rate) significant at specified levels of significance and also (a) within fifty markers of, or (b) on the same chromosome as a major gene was calculated. For purpose (ii), scan statistics of length 15 were calculated for each chromosome and their empirical significance levels estimated on the basis of 500 replicates generated under no linkage. The scan statistic was applied to the mhlod scores from one replicate (Replicate 5). Empirical p-values for the scan statistic were determined by computing mhlod scores on 500 replicates of simulated null data. For purpose (i), significance levels between 0.001 and 0.01 had the greatest hit rate for all five methods and both criteria. For criterion (a) at the 0.001 level of significance, both mlod and mhlod displayed the highest hit rates, approximately 0.4 for each. For criterion (b), all methods but ALL_SIBS and ASP had hit rates ranging between 0.4 and 0.5. For purpose (ii), the scan statistic proved equally or more powerful than the single-locus statistic for two of the seven susceptibility genes while the remaining five genes were not detected.

Chromosome Mapping↗
PubMed1994

Statistics in Medicine: citations of papers in the first ten years.

All papers from Volume 1 of Statistics in Medicine were followed up in the Science Citation Index. There were 6.7 citations per paper in medical journals as opposed to 1.5 citations per paper in statistical journals, and overall there were 8.7 citations per paper. The average citation rate was lower than that of the first issue of Biometrics, published in the same year, but this was partly because of a greater proportion of zero-cited papers. Citations from medical journals increased annually, until about 5 years after publication, after which they remained steady. Volumes 2-6 of Statistics in Medicine and Volumes 29-33 of Biometrics, for the years 1983-87, were followed up for statistical/medical source of the citations. Citations for Volumes 2-4 of Statistics in Medicine were relatively low, but picked up by Volume 5. Biometrics had more citations from statistical sources than from medical, by contrast to Statistics in Medicine which had far more citations from medical sources than from statistical.

Clinical Trials as Topic↗
PubMed1986

Learning statistical methods.

Since at least 1951 anesthesia journals have called for a more rigorous application of statistical methods in research reports. This appeal for statistical excellence actually applies to the researcher, to the clinician journal reader, and to the editor. Thirty five years ago the obligations of these three groups was made clear in an unsigned editorial in Anesthesiology; there is now a widespread consensus on these responsibilities. The researcher must create valid science. The clinician reader must bring sufficient intellectual skills to understand a journal article; using these skills the reader must critique the research report to judge its applicability to his patients. The editor must decline manuscripts showing poor or absent application of the scientific method, experimental design, and statistical analysis. Though the editors continue to exhort further improvements, even a casual perusal of their journals demonstrates a tremendous improvement in the handling of numeric data over the last four decades. With the increasing sophistication of statistical methods in journals, the reader must continue to expand his statistical understanding. In this short review, a few highlights of statistical methods useful either in planning and accomplishing a research project or in reading a research report will be discussed; these include the planning of a study design, data collection, data analysis, and interpretation of the research. Comments about using these concepts to better understand a research article will be included. Also included will be an annotated reading list for further study and reference. Mathematical formulas have been avoided as much as possible. Actual use of statistics requires use of equations which can be found in the books of the reading list.

Biometry↗
PubMed2006

[Expansion or reduction of geriatric care structures in Germany? A critical analysis on the significance of the official statistics and other surveys].

The legal survey basis for the hospital statistics of the Statistisches Bundesamt (German Federal Statistical Office) affecting the recording of data starting in the year 2002 has been also adjusted to improve the quality of information on geriatric care structures. The basic hospital statistics data for the year 2003 published in April 2005 report 171 geriatric hospital facilities for in-house treatment and 97 for partial in-house treatment as well as 74 geriatric rehabilitation facilities for inpatient treatment. In an additional internal investigation, another 46 geriatric rehabilitation facilities for outpatient treatment were ascertained for the year 2003. Compared to other, earlier surveys, the Statistisches Bundesamt reports an almost equal number of geriatric care facilities in the hospital sector, but a far lower number of such facilities in the sector of rehabilitation facilities for inpatient treatment, and therefore is highly incomplete. Hence, despite modified recording conditions, the official statistics do not provide a realistic representation of geriatric care structures. Under consideration of these limitations and corresponding corrections, the average geriatric care ratio (inpatient and partial inpatient or out-patient geriatric treatment places in hospitals and rehabilitation facilities per 10,000 persons aged 65 and above) amounted to 10.2 geriatric treatment units in 1997, 12.2 in 2000, and 12.3 in 2003. There were significant differences regarding the total capacity and the shares of different kinds of geriatric care structures in the individual federal states. All in all, that means that the expansion of geriatrics that had taken place until the year 2000 has slowed down significantly over recent years and has largely been limited to demographic adjustments. As far as the relevance of reliable numbers on the existence of geriatric care structures for requirement planning, secondary statistics, and state-related comparative analyses is concerned, the ongoing weaknesses of the hospital statistics must be taken into account when developing corresponding interpretations. They call for examining further improvements of the procedure of recording geriatric facilities for the official statistics. So far, they are no reliable basis for cross-sectional analysis.

Aged↗
PubMedDate not supplied

[Exactness of mortality statistics by external and natural causes of death with medico-legal intervention in Catalonia, 1996].

OBJECTIVE: Deaths due to external and natural causes with forensic intervention require medical-legal autopsy. In order to be included in the mortality statistics the results are reported in the statistical document MNP52. The accuracy of cause of death depends on the characteristics of the document, the point at which it is completed (after the death or after the autopsy) and the person that completes it. The objective is to determine the accuracy of external and natural causes of death with forensic intervention, reported in the official statistical documents by a medical-legal autopsy report of these deaths occurred in Catalonia in 1996. METHODS: Two samplings were undertaken--one for natural causes and another for external causes--that were stratified by sex and judicial district. The information sources were the Mortality Register of Catalonia for the statistical documents and the criminal courts for the medical-legal autopsy, toxicological and pathological reports. We calculated the index of agreement, the sensitivity or detection rate (DR) and the positive predictive value of confirmation rate (CR), and their respective 95% confidence intervals. RESULTS: The index of agreement was 72.3% (IC 95%: 68.7-75.9). The DR for external causes groups was 65.9% (60.6-71.2) and the CR was 69% (63.6-71.2). For natural causes the DR was 79.4% (74.7-84.2) and the CR was 75.5% (70.7-80.5). CONCLUSION: In deaths with forensic intervention, the official statistical documents do not correctly report external causes of death, and statistics for natural causes of death approach acceptable levels of accuracy. The results are mainly due to deficits in reporting and certifying these causes in the official statistics.

Algorithms↗
PubMed2000

Statistics of transmitter release at nerve terminals.

This review presents an historical account of the developments of the statistical analysis of quantal transmission over the past half century and of the progress made in using this approach to reveal new properties of nerve terminals. In the early 1950s, Katz and his colleagues showed that evoked transmitter release occurred in quanta at the neuromuscular junction, opening up the study of transmitter release at nerve terminals to statistical analysis. In the subsequent two decades attempts were made to see if evoked quantal release could be described by binomial or compound binomial statistics, as originally suggested by Katz, and to relate the parameters of the statistic to various structures of the nerve terminal. During this period two hypotheses were enunciated, namely the 'vesicle hypothesis', which states that quanta arise as a consequence of the packaging of transmitter in vesicles; and the 'active zone hypothesis', which states that vesicles undergo exocytosis at discrete sites on the nerve terminal. Unsuccessful attempts were made to relate the binomial parameter n to the elements in these hypotheses, that is to the number of active zones possessed by the terminal or the number of vesicles available for release at these zones. This difficulty was part resolved in the late 1970s with the application of non-uniform binomial statistics to transmitter release from nerve terminals, in which n is the number of active zones each with their individual probabilities, p(j). Autocorrelation functions were subsequently introduced to detect if transmitter release is quantised at a particular nerve terminal. Statistical methods which would allow discrimination between different models of transmitter release over the active zones of a terminal were then developed. The introduction of maximum likelihood estimation procedures then allowed estimates to be made of the parameters in the statistical models of quantal release. The application of these procedures to experimental data from a variety of nerve terminals provided evidence for the concept that each synapse, taken as possessing a single active zone, possesses its own individual probability of secretion of a quantum by the exocytosis of a vesicle. In the late 1960s Stevens introduced the first stochastic approach to the analysis of the kinetics of the release of a quantum of transmitter at the neuromuscular junction following an impulse. In the subsequent decades this was developed into an explicit theory for the interaction of proteins involved in regulated exocytosis of a vesicle at an active zone. The parameters were the number of transition steps in the release process (k), each occurring at the same rate (alpha), with the possibility of each of these steps becoming blocked at the same rate (gamma). Maximum likelihood estimation procedures could then be used to obtain these parameter values. The discovery was made in the 1990s of the core proteins of the SNARE complex that govern regulated exocytosis. This offers the possibility in the near future of identifying the kinetic interaction of these proteins with the parameters of the stochastic process of exocytosis which confer a particular probability on individual synapses.

Animals↗