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.

SourceResearch connectionFrom yearThrough year

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

At least 847 records · Page 47Linked to original sources

PubMed2004

Statistical validation of surrogate endpoints: is bone density a valid surrogate for fracture?

In the treatment of osteoporosis using anti-resorptive agents there has been increasing interest in quantifying the relationship between fracture endpoints and surrogates such as bone mineral density (BMD) or bone turnover markers. Statistical methodology constitutes a critical component of assessing surrogate validity. Depending on study designs, data resources, and statistical methods used for analyses, one has to use caution when interpreting results from different analyses, especially when results are disparate. For example, analyses based on individual patient data reported that only a limited proportion of the anti-fracture efficacy was explained by BMD increases for agents such as alendronate, risedronate and raloxifene. Analyses employing meta-regression based on summary statistics, however, indicated that most of the anti-fracture benefits were due to improvements in BMD. In this paper, we review definitions of surrogate endpoints and requirements for their statistical validation. We evaluate whether BMD meets these requirements as a possible surrogate for fracture. Our review indicates that the actual BMD value is correlated with fracture risk and thus BMD is useful in identifying patients that might need treatment. There is limited evidence to support BMD increase with anti-resorptive agents as a reliable substitute for fracture risk reduction. Strengths and limitations for various statistical methods are discussed.

Biomarkers↗

PubMed2005

Measuring statistical literacy.

This study considers the measurement of Statistical Literacy understanding that goes beyond the basic chance and data skills and knowledge in the mathematics curriculum. This understanding requires application of mathematical skills in a range of contextual situations and draws on aspects of statistics, such as variation and inference, which may not be explicit in the school curriculum. The study reports the outcomes from tests of Statistical Literacy given to 673 students from Grades 5 to 10. It confirms the nature and structure of a previously identified construct of Statistical Literacy and proposes three subgroups of items that address aspects of Statistical Literacy that might usefully be measured by classroom teachers.

Adolescent↗

PubMedDate not supplied

Knowledge of statistical methods and their implications for clinical practice: a survey of paediatricians.

Improvements in computer software have contributed to an increase in the use of multivariate statistical analyses e.g. multipLe regression in recent times. Our aim was to assess the familiarity with, and understanding of these complex statistical methods among Irish paediatricians. Questionnaires were sent to all paediatric specialist registrars (SpRs) and consultants in the Republic of Ireland. The questionnaire detailed information about clinical practice, and contained a short quiz on statistical methods. 137 questionnaires were distributed, and 62 (45.6%) were returned. Eighty four percent of respondents aimed to read journals weekly but only 46.7% managed to. The commonest journals used were; Archives of Disease in Childhood (93%), Pediatrics (53%), British Medical Journal (46%) and Journal of Paediatrics (45%). 28 of 61 (45.9%) of respondents have had further training in statistics. Only 19% felt they had a clear understanding of regression. Fifty-eight of 62 respondents (93.5%) completed the short test. The average score was 5.1/10. Sixty seven percent of questions on basics were answered correctly, 37.9% of questions on application of tests were answered correctly and 25.8% of questions on regression were answered correctly. Only 3.4% answered all questions on regression correctly. The overall knowledge of advanced statistical methods was poor. There is a poor overall understanding of the concept of regression, despite its increasingly common use.

Health Knowledge, Attitudes, Practice↗

PubMed1992

The state of federal health statistics on racial and ethnic groups.

OBJECTIVE: To examine assumptions underlying federal health statistics on racial and ethnic groups in the United States. DATA SOURCES: Studies conducted by federal agencies and other investigators, and technical appendices of published vital statistics and census reports. DATA SYNTHESIS: Several assumptions underlying federal health statistics on racial and ethnic groups are not well supported. Conceptual (as opposed to operational) definitions of race and ethnicity are not available, and scientific grounds for definition are not considered. Procedures for the ascertainment of race and ethnicity vary within and among data-collection agencies. Miscounting and misclassification may vary by an order of magnitude between whites and other races. The responses of individuals to questions of racial and ethnic identity differ for different indicators, in different surveys, and at different times. As a result, counts, rates, and rate ratios may not be meaningful or accurate. Particularly for Hispanics and for races other than whites or blacks, there are inconsistencies in statistical information that may hinder health research and program development. CONCLUSIONS: Improvement of federal health statistics for racial and ethnic groups requires (1) clarification of goals for classification, (2) adoption of scientific principles for the validation and definition of the categories "race" and "ethnicity," (3) assessment of perceived social identity in the population, and (4) periodic evaluation.

Data Collection↗

PubMed1991

[The importance of ADP for medical statistics].

Recent german articles concerning medical statistics give priority to data acquisition. This contribution describe the statistical analysis and presentation of the results of the Swiss accident insurance Company (SUVA) with 1.9 million of insured workers. This documentation covers per year approximately 450,000 accidents at work and in free time. The documentation include primarily the insurance-data, for medical purposes we use the adjacent diagnoses (up to 20 per accident). The new medical statistic (SUMEST) begins with the accidents of the year 1984, updated until 1988, this means, that we have a statistic of the outcome of accidents. For the analysis we use properties of the personality, details of what happened, injury, expenses for healing and results. The primary presentation is a collection of tables for every diagnosis (total 417) with the most important criteria for the cases. It is provided to publish results of special problems like time of disability, influence of algodystrophy etc. The possibility to analyse a big documentation (25 MBytes) with a personal computer give to the physician the possibility to perform the statistical analysis himself and to select the significant results immediately.

Data Interpretation, Statistical↗

PubMed1991

An approach to select the appropriate statistical method for testing bioequivalence.

Since most bioavailability studies are usually done with only a limited number of volunteers (usually 10-30), the statistical properties of the calculated bioavailability parameters are not well defined. The established statistical methods to test bioequivalence are usually based on either the assumption of normality or a symmetrical distribution. However, the decision of which method to apply, depends primarily on the distributional assumption of the data. In this study, an approach is followed where the small data base of a limited number of volunteers is expanded by adding pseudo-volunteers by "bootstrap" simulations. From such a larger data base it is easier to determine the statistical distributional properties of bioavailability parameters, which in its turn leads to the identification of an appropriate statistical method. With more certainty on which statistical method to apply, the original data can be used more effectively in testing for bioequivalence. In this paper, comparisons are made between the distributions of bioavailability parameters of an actual 60-volunteer study and those of two simulated data sets. Each such data set contained a random sample of 10 volunteers each (from the 60 volunteers), together with 50 pseudo-volunteers. These 50 volunteers were simulated from the random sample of 10 real volunteers. Good correspondences were obtained when comparing these two data sets with the original data, which indicated the validity to use this approach in bioavailability studies where a small number of volunteers had been used. This method proved useful to define the distributional properties for a relative small number of parameter-values available.

Biological Availability↗

PubMed1987

Statistical techniques reported in pathology journals during 1983-1985. Implications for pathology educators.

The increasingly scientific basis of medicine challenges pathology educators to incorporate quantitative skills into a very compressed curriculum. One strategy is to focus training on the statistical techniques pathologists will most commonly encounter in their literature. We identified the reporting of statistical techniques in nearly 5200 original articles published during 1983-1985 in 16 journals important to pathology. Our results suggest that a reader familiar with 12 fundamental statistical concepts can evaluate knowledgeably over 95% of the quantitative findings reported in these journals. With the exception of survival analysis and pharmacologic modeling, these techniques are typically encountered in many introductory statistical texts. In numerous articles, failure to identify the statistical methodology used made it impossible to identify the analytic procedures used and, hence, judge the scientific validity of results.

Curriculum↗

PubMed1985

Statistical and research quality of the medical and pharmacy literature.

The statistical and research quality of reports published in two U.S. medical journals and two U.S. pharmacy journals over a 12-month period was determined. A 50% random sample of issues of New England Journal of Medicine, Annals of Internal Medicine, American Journal of Hospital Pharmacy, and Drug Intelligence and Clinical Pharmacy published in 1979 was reviewed, and all citable items were classified as one of nine types of communications. Items classified as original evaluative research reports were evaluated for experimental design and research goals and rated for appropriateness of statistical testing and overall research quality according to a set of objective criteria. Reports that were not immediately classifiable were reviewed by one additional person, and classifications or ratings were assigned by mutual consent. Of a total of 1506 citable items, 120 met the criteria for original evaluative research reports. The two medical journals had the highest percentages of reports for which statistical methods were rated as correct but also had the most reports for which statistical methods could not be rated as a result of incomplete documentation or publication errors. Reports in the medical journals had conclusions based on a logical progression of hypothesis, methods, and analysis of results more frequently than did reports in pharmacy journals. Reports in the medical and pharmacy journals differed greatly according to research design and research goals. Improvement is needed in the statistical quality and research quality of original research reports published in the pharmacy literature.

Pharmacy↗

PubMed1979

Some non-random views of statistical significance.

Although research in family medicine is growing rapidly, few family physicians have had experience or training in statistical methods. Statistical significance and P values are often misunderstood and frequently misapplied. "Significance" is arbitrary; the actual P value is of greater interest than a significant/not significant statement; P values do not measure the strength of an association; statistical significance is not equivalent to "actual" significance; P values are largely dependent on sample size; and data "dredging" is guaranteed to yield spurious results. Competent statistical consultation, careful study planning, and recognition of statistical pitfalls are important to anyone who does research, and knowledge of these areas is useful as well to anyone reading the medical literature.

Family Practice↗

PubMed1983

Validity and reliability of trends in suicide statistics.

For a number of reasons, suicide is under-reported and the reliability of official rates is subject to error through variations in defining and reporting cases, the kind of inaccuracy encountered when ascertaining cases in studies of mortality from any cause. Nevertheless, the evidence from studies designed to see whether these sources of error invalidate the differences reported between cultural and social groups indicates that they are of a random nature, at least to an extent that allows epidemiologists to compare rates between countries and districts within them, between demographic groups, and over periods of time. The accuracy and hence the value of official suicide statistics has been questioned in recent years to an extent that has led some authorities to dismiss their usefulness in epidemiological research. In England and Wales, a decision as to whether an unexpected or violent death is a suicide or not is normally made at a coroner's inquest; similar, but by no means identical procedures are followed in other developed countries. Cases defined and ascertained in this way are the usual source of national suicide statistics. Indeed, it may be claimed that the medical and post-mortem inquiry together with the evidence of witnesses on the psychological and social circumstances relating to the act, entail a more thorough investigation into this cause of death than is usual in compiling the statistics of deaths from other causes. Consider by contrast the General Registrar Office enquiry into the accuracy of reporting other causes of death (1). When, for example, deaths from cancer of the lung are registered before holding a post-mortem and the cause of death is later checked by a pathologist, the net error is about 16%. The conclusions drawn from official cancer statistics, however, are not seriously questioned. It is thus reasonable to expect suicide statistics, despite inaccuracies, also to be of heuristic value particularly as over-reporting is negligible, and it is only under-reporting, not a matter in dispute, that need be considered. Studies of under-reporting of suicide include those done in England and Wales (2), Scotland (3), and in Ireland where McCarthy & Walsh (4) examined coroners' case records in Dublin using clinical criteria to assess the probability of suicide. Though their revised rate was considerably higher than the official one, the Irish rate is still exceptionally low when compared with other countries. Barraclough (5) confirmed by other means that the adjusted rate was still half that of England.(ABSTRACT TRUNCATED AT 400 WORDS)

Adolescent↗

PubMed1982

An alternative approach to teaching statistics to dental students.

From a review of the literature, instruction in statistics appears to be common in dental schools but beset with problems. Based, in part, on several surveys of the statistics content of dental literature, there is some general agreement that dentists-in-training need to acquire a working knowledge of statistics. Content guidelines developed for health science statistics courses are available and are considered entirely appropriate for dental education. Several authors have suggested that computer-assisted instruction is ideally suited to minimize the problems encountered in teaching statistics to dental students. Computer-assisted instruction is expensive, however, and self-instruction using a programmed text was found, on the basis of preliminary evaluation, to be an effective and inexpensive alternative.

Curriculum↗

PubMed1995

[Utilization++ of statistical analysis in original articles published in the Revista Clínica Española (1992-1993)].

OBJECTIVES: To report the types of statistical analysis performed and to quantitate statistical accessibility in the originals of Revista Clínica Española from 1992 to 1992. METHODS: Ninety-one originals were reviewed. Statistical analysis were classified according to 18 categories and statistical accessibility was evaluated from the same categorization. RESULTS: Seventy-four percent of originals used inferential methods. The most frequently used categories were Bivariable tables (46%) and t and z test (42%). Thirty-six percent of originals used some multivariable analysis and particularly the Analysis of survival (12%). CONCLUSIONS: Considering the high frequency of use of multivariable techniques, investigators need to have some knowledge on statistics beyond a basic course.

Periodicals as Topic↗

PubMed1995

Practical epidemiology and biostatistics in research. Study design and statistics.

Soundly planned study may well lead to the findings that are of wide scientific application and interest. This paper is intended to provide a simple and systematic guideline pertinent to the design, analysis and interpretation of studies, especially in the health care. It emphases on the importance of 'Statistics' in the design, conduct, analysis and interpretation of the studies. It is pointed out that all the stages of research studies are vulnerable to statistical mismanagement. The most important thing is to balance the interests of the individuals in the study with those of the much larger number who may benefit in the long run. Concerning the proper use of statistics, the following things are recommended. (a) Statistical advice is needed at the planning stage of a project and not at its end; (b) statistical advice may be saught at almost every stage of a project; (c) the important aspects of logic and correctness of argument may be carefully looked before interpreting the results and (d) finally, before getting any work published, opinion of an expert statistician must be saught, so that erroneous conclusions may not become enshrined as the truth.

Data Collection↗

PubMed1994

Transfer of technology from statistical journals to the biomedical literature. Past trends and future predictions.

OBJECTIVE: To investigate the speed of the transfer of new statistical methods into the medical literature and, on the basis of current data, to predict what methods medical journal editors should expect to see in the next decade. DESIGN: Influential statistical articles were identified and the time pattern of citations in the medical literature was ascertained. In addition, longitudinal studies of the statistical content of articles in medical journals were reviewed. MAIN OUTCOME MEASURES: Cumulative number of citations in medical journals of each article in the years after publication. RESULTS: Annual citations show some evidence of decreasing lag times between the introduction of new statistical methods and their appearance in medical journals. Newer technical innovations still typically take 4 to 6 years before they achieve 25 citations in the medical literature. Few methodological advances of the 1980s seem yet to have been widely cited in medical journals. Longitudinal studies indicate a large increase in the use of more complex statistical methods. CONCLUSIONS: Time trends suggest that technology diffusion has speeded up during the last 30 years, although there is still a lag of several years before medical citations begin to accrue. Journals should expect to see more articles using increasingly sophisticated methods. Medical journals may need to modify reviewing procedures to deal with articles using these complex new methods.

Forecasting↗

PubMed1994

Comments on the statistical aspects of the NRC's report on DNA typing.

The goal of the NRC report on DNA typing was to answer a "crescendo of questions concerning DNA typing," many of them in the areas of population genetics and statistics. Unfortunately, few of these questions were answered adequately. In lieu of answering these questions, the panel proposed another conservative method of forensic inference, the "ceiling principle." Aside from its extreme conservativeness, this new method is difficult to justify because it is based on inadequate population genetics and statistical theory. Moreover, in its ultimate implementation, the panel's method will depend on a population genetics study whose rationale is questionable. In this article, we elaborate some of the general comments we made about the NRC report in a recent article [1]. Specifically we cover three topics. First we question the statistical basis for the ceiling principle, showing that the empirical results that motivated the method are likely to be misinterpreted and showing, by power calculations, that the effects of population substructure cannot be substantial. Second, we show that the study design to determine "ceiling" allele frequencies has several undesirable statistical properties. Finally, we discuss the estimation of handling errors from the statistical perspective, a subject treated inadequately by the report.

DNA Fingerprinting↗

PubMedDate not supplied

[Medicine and statistics].

Doctors as researchers often resort to hypothesis testing and confidence intervals in the statistical analysis of study results. On the one hand, the use of statistical software has made this task easier. On the other hand, this has not contributed to neither a better understanding of statistical concepts nor to a seminal cooperation between doctors and statisticians. This note shows that the classical statistical theory, also called frequentist or Fisherian, offers probability statements that are very indirect ones. Bayesian statistics, on the other hand, gives a probability statement which is a more direct answer to the research question. Of course, this latter approach would hardly be considered without a close cooperation with an statistician. An improvement in the meaning of the probability statements may be seen as a gain in quality. An effective cooperation with statisticians, an issue that is more ethical than methodological, is directed to achieving a research of a better quality, so avoiding the scandal of poor medical research.

Humans↗

PubMedDate not supplied

[Statistical results: which method of presentation to chose?].

Hypothesis testing and significance is currently the most widely used method in the medical literature to report statistical results. However, this method has several limitations. The main one is linked to the risk of misinterpretation of the p value. The arbitrariness of the 5 percent value used to determine whether a result is or not statistically significant is not always kept in mind, and the concept of statistical significance might therefore be confused with that of clinical or biological relevance. The misinterpretation pitfalls are mostly linked to the fact that the p value does not give precise indications on the strength of the association and its direction, or on the variability in the sample. Therefore, some experts claim that hypothesis testing and significance should be avoided in reporting statistical results, and that the method based upon estimation and confidence interval should be more widely used. By this latter method, it is possible to know the direction of the association and the effect size (i.e. the strength of the association). The precision of the estimation, i.e. the variability of the estimation in the sample, can be assessed by the width of the confidence interval: the narrower the confidence interval, the more precise the estimation. Therefore, the clinical relevance of the findings is easier to infere from such results than from those only reporting p values. However, the estimation and confidence interval method is not without its own limitations. This method is difficult to apply to non-parametric tests, and for some results, such as the comparison of mortality ratios, the p value is highly informative. On the other hand, the misinterpretation risk is not totally ruled out when estimation and confidence interval method is used. In the situations where both methods can be employed, there is not yet in the scientific community a definite consensus on which method is the best one to report statistical results, hence some experts suggest that both methods can be presented simultaneously, especially for clinical and epidemiological studies.

Confidence Intervals↗

PubMed1976

Annual summary of vital statistics-1975.

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, both published by the Statistical Office of the United Nations, and the World Health Statistics Report, published by the World Health Organization. All the United States data for 1975 are estimates based upon a 10% sample of material 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; state information should be interpreted cautiously.

Birth Rate↗