Opinion poll on WHO health statistics publications: morbidity statistics.
Explore the source record for details and available documents.
Communicable Diseases↗
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.
[National statistics on mediastinal neoplasms. 1. Compilation of statistics on mediastinal neoplasms].
Explore the source record for details and available documents.
Humans↗
[National statistics on mediastinal neoplasms. 2. National statistics on the clinical characteristics of thymoma].
Explore the source record for details and available documents.
Adolescent↗
Use of statistical analysis in the AJR and Radiology: frequency, methods, and subspecialty differences.
OBJECTIVE: The statistical procedures appearing in Radiology and the AJR were analyzed to determine the types of statistical tests used in major articles, their relative frequencies, and whether these methods differed according to radiologic subspecialty. MATERIALS AND METHODS: Six hundred sixty-nine major articles published in Radiology or the AJR during 1993 were reviewed for statistical content. These articles were first classified by radiologic subspecialty or principal focus according to the named journal section where each appeared (e.g., breast imaging, musculoskeletal radiology, neuroradiology). The statistical methods used in each article were then assigned to one of 18 categories (e.g., descriptive, basic t- and z-tests, correlation/regression techniques, and nonparametric statistics). RESULTS: Of the 669 major articles analyzed, 294 (44%) used no statistical methods or descriptive statistics only, 179 (27%) used only one type of statistical method, 102 (15%) used two methods, and 94 (14%) used three or more methods. Statistical techniques were most likely to be used in articles classified as cardiac (85% used statistical analysis) or contrast agents (79%), but least likely to be used in gastrointestinal/abdominal imaging (45%), radiotherapy (44%), or miscellaneous topics (29%). Five basic categories of statistical methods--basic z- and t-tests, basic decision statistics (sensitivity, specificity), basic contingency table analysis (chi 2, McNemar), correlation/regression techniques (linear regression, Pearson coefficients), and basic nonparametric tests (Wilcoxon, Mann-Whitney U)--account for the complete statistical analysis appearing in approximately 80% of the major articles. CONCLUSION: Approximately half the major articles appearing in Radiology or AJR used no statistics, descriptive analysis only, or a simple inferential test employing t-tests or confidence intervals; advanced statistical techniques were found in only 20% of articles. Statistical analysis was most likely to be used in articles dealing with cardiac radiology or contrast agents and least likely to be used in articles on abdominal/gastrointestinal radiology, radiotherapy, and miscellaneous radiologic topics.
Humans↗
Review of statistics usage in the American Journal of Obstetrics and Gynecology.
OBJECTIVE: Our purpose was an assessment of statistical analysis in studies published in the American Journal of Obstetrics and Gynecology, as well as documentation of appropriate and inappropriate statistical application. STUDY DESIGN: All papers included in the Clinical Articles section and transactions of societies sections of the January through June 1994 issues of the American Journal of Obstetrics and Gynecology (volume 170, numbers 1 to 6) were reviewed for statistical usage. Each paper was given a rating for the thoroughness of the listing of applied statistics and a rating for the appropriateness of statistical usage, when possible. RESULTS: Of the 190 available articles, 53 consisted of studies void of statistics, 8 of which required statistics or claimed significance without the use of statistics. Therefore 145 articles were included in the final analysis. Because of inappropriate or incomplete descriptions of statistics used within the article (52.6%), the ability to assess the appropriateness of usage was severely limited. However, 44 articles (30.3%) could be classified as having appropriate usage of statistics, whereas 46 articles (31.7%) were deemed to have inappropriate usage of statistics. Furthermore, 27 of these 46 articles were noted to have serious flaws. CONCLUSION: The lack of complete and detailed listings of applied statistics made it difficult to assess the appropriateness of more than half the studies examined, suggesting a need for more detailed guidelines as to the listing of statistical procedures used. Despite this fact, nearly one third of the articles contained examples of statistics used inappropriately. These findings suggest that a policy of statistical review be instituted.
Publishing↗
Nonparametric analysis of statistic images from functional mapping experiments.
The analysis of functional mapping experiments in positron emission tomography involves the formation of images displaying the values of a suitable statistic, summarising the evidence in the data for a particular effect at each voxel. These statistic images must then be scrutinised to locate regions showing statistically significant effects. The methods most commonly used are parametric, assuming a particular form of probability distribution for the voxel values in the statistic image. Scientific hypotheses, formulated in terms of parameters describing these distributions, are then tested on the basis of the assumptions. Images of statistics are usually considered as lattice representations of continuous random fields. These are more amenable to statistical analysis. There are various shortcomings associated with these methods of analysis. The many assumptions and approximations involved may not be true. The low numbers of subjects and scans, in typical experiments, lead to noisy statistic images with low degrees of freedom, which are not well approximated by continuous random fields. Thus, the methods are only approximately valid at best and are most suspect in single-subject studies. In contrast to the existing methods, we present a nonparametric approach to significance testing for statistic images from activation studies. Formal assumptions are replaced by a computationally expensive approach. In a simple rest-activation study, if there is really no activation effect, the labelling of the scans as "active" or "rest" is artificial, and a statistic image formed with some other labelling is as likely as the observed one. Thus, considering all possible relabellings, a p value can be computed for any suitable statistic describing the statistic image. Consideration of the maximal statistic leads to a simple nonparametric single-threshold test. This randomisation test relies only on minimal assumptions about the design of the experiment, is (almost) exact, with Type I error (almost) exactly that specified, and hence is always valid. The absence of distributional assumptions permits the consideration of a wide range of test statistics, for instance, "pseudo" t statistic images formed with smoothed variance images. The approach presented extends easily to other paradigms, permitting nonparametric analysis of most functional mapping experiments. When the assumptions of the parametric methods are true, these new nonparametric methods, at worst, provide for their validation. When the assumptions of the parametric methods are dubious, the nonparametric methods provide the only analysis that can be guaranteed valid and exact.
Algorithms↗
Construction of null statistics in permutation-based multiple testing for multi-factorial microarray experiments.
MOTIVATION: The parametric F-test has been widely used in the analysis of factorial microarray experiments to assess treatment effects. However, the normality assumption is often untenable for microarray experiments with small replications. Therefore, permutation-based methods are called for help to assess the statistical significance. The distribution of the F-statistics across all the genes on the array can be regarded as a mixture distribution with a proportion of statistics generated from the null distribution of no differential gene expression whereas the other proportion of statistics generated from the alternative distribution of genes differentially expressed. This results in the fact that the permutation distribution of the F-statistics may not approximate well to the true null distribution of the F-statistics. Therefore, the construction of a proper null statistic to better approximate the null distribution of F-statistic is of great importance to the permutation-based multiple testing in microarray data analysis. RESULTS: In this paper, we extend the ideas of constructing null statistics based on pairwise differences to neglect the treatment effects from the two-sample comparison problem to the multifactorial balanced or unbalanced microarray experiments. A null statistic based on a subpartition method is proposed and its distribution is employed to approximate the null distribution of the F-statistic. The proposed null statistic is able to accommodate unbalance in the design and is also corrected for the undue correlation between its numerator and denominator. In the simulation studies and real biological data analysis, the number of true positives and the false discovery rate (FDR) of the proposed null statistic are compared with those of the permutated version of the F-statistic. It has been shown that our proposed method has a better control of the FDRs and a higher power than the standard permutation method to detect differentially expressed genes because of the better approximated tail probabilities.
Algorithms↗
Statistical reviewing policies in dermatology journals: results of a questionnaire survey of editors.
BACKGROUND: Problems with statistical methods and reporting have been noted in articles published in dermatology journals. Conclusions presented in published reports may be misleading if based on inappropriate or misinterpreted statistical analysis. OBJECTIVE: We sought to assess dermatology journal editors' policies and perceptions regarding statistical review of submitted manuscripts. DESIGN: We mailed and e-mailed a questionnaire survey. PARTICIPANTS: A total of 43 dermatology journal editors, representing 35 dermatology journals from the United States and abroad, participated in this study. RESULTS: In all, 32 editors (74.4%), representing 30 journals (85.7%), returned questionnaires. A total of 24 editors (75%) reported having requested statistical reviews on less than 5% of published manuscripts containing original quantitative analysis (ie, excluding reviews and case reports), whereas 3 editors (9.4%) reported having requested statistical reviews on more than 75% of such manuscripts. Most editors reported requesting statistical reviews on a case-by-case basis either after initial favorable review by subject-matter (nonstatistical) reviewers (12 editors; 37.5%) or at the same time that subject-matter review was requested (6 editors; 18.8%). A total of 4 editors (12.5%) reported requesting statistical review for all manuscripts at the same time they are sent for subject-matter review. Another 10 editors (31.3%) said their journals had no general policy on statistical reviewing, and statistical review is almost never needed. For 15 editors (46.9%), ideal statistical reviewing policy was identical to their current policy, whereas 13 (40.6%) favored a more rigorous and 3 (9.4%) a less rigorous policy. CONCLUSIONS: Dermatology journals infrequently perform statistical reviews of submitted manuscripts. Dermatology journal editors' statistical review policies range from no general policy to (most frequently) requesting reviews on a case-by-case basis to reviewing all submitted manuscripts. Many editors favor more rigorous statistical reviewing policies for their journals. Increased use of statistical reviewing may increase the reliability of conclusions published in dermatology journals.
Dermatology↗
Statistical reviewing policies of medical journals: caveat lector?
OBJECTIVE: To describe the current policies regarding statistical review of clinical research in biomedical journals. DESIGN: Cross-sectional survey. PARTICIPANTS: Editors of biomedical journals that publish original clinical research. MEASUREMENTS: General policies on statistical review, types of persons used for statistical reviewing, compensation of statistical reviewers, percentage of articles subject to such review, percentage of time statistical review makes an important difference, journal circulation, and selectivity. MAIN RESULTS: Of 171 journals, 114 (67%) responded to the survey. About one third of journals had policies that guaranteed statistical review for all accepted manuscripts. In approximately half of the journals, articles were sent for statistical review at the discretion of the editor. There was some evidence that statistical review policies differed between journals of different circulation size. In journals in the top quartile of circulation (> 25,000) the probability of definitely having a statistical review before an acceptance decision was 52%, but it was only 27% in journals in the lower three quartiles (p = .09). The probability of a statistical consultant on staff ranged from 31% in the bottom quarter, to 58% in the middle two, to 82% in the highest quarter (p < .001). Editors judged that statistical review resulted in an important change in a manuscript about half of the time. CONCLUSIONS: Except in the largest circulation medical journals, the probability of formal methodologic review of original clinical research is fairly low. As readers and researchers depend on the journals to assess the validity of the statistical methods and logic used in published reports, this is potentially a serious problem. This situation may exist because the cost of such statistical review can be considerable, and because finding appropriate reviewers can be difficult. It may also exist partly because editors or publishers may not regard such review as important. The professions of medical publishing, statistics, epidemiology, and other quantitative disciplines should work together to address this problem.
Cross-Sectional Studies↗
Statistical assessment of the learning curves of health technologies.
OBJECTIVES: (1) To describe systematically studies that directly assessed the learning curve effect of health technologies. (2) Systematically to identify 'novel' statistical techniques applied to learning curve data in other fields, such as psychology and manufacturing. (3) To test these statistical techniques in data sets from studies of varying designs to assess health technologies in which learning curve effects are known to exist. METHODS - STUDY SELECTION (HEALTH TECHNOLOGY ASSESSMENT LITERATURE REVIEW): For a study to be included, it had to include a formal analysis of the learning curve of a health technology using a graphical, tabular or statistical technique. METHODS - STUDY SELECTION (NON-HEALTH TECHNOLOGY ASSESSMENT LITERATURE SEARCH): For a study to be included, it had to include a formal assessment of a learning curve using a statistical technique that had not been identified in the previous search. METHODS - DATA SOURCES: Six clinical and 16 non-clinical biomedical databases were searched. A limited amount of handsearching and scanning of reference lists was also undertaken. METHODS - DATA EXTRACTION (HEALTH TECHNOLOGY ASSESSMENT LITERATURE REVIEW): A number of study characteristics were abstracted from the papers such as study design, study size, number of operators and the statistical method used. METHODS - DATA EXTRACTION (NON-HEALTH TECHNOLOGY ASSESSMENT LITERATURE SEARCH): The new statistical techniques identified were categorised into four subgroups of increasing complexity: exploratory data analysis; simple series data analysis; complex data structure analysis, generic techniques. METHODS - TESTING OF STATISTICAL METHODS: Some of the statistical methods identified in the systematic searches for single (simple) operator series data and for multiple (complex) operator series data were illustrated and explored using three data sets. The first was a case series of 190 consecutive laparoscopic fundoplication procedures performed by a single surgeon; the second was a case series of consecutive laparoscopic cholecystectomy procedures performed by ten surgeons; the third was randomised trial data derived from the laparoscopic procedure arm of a multicentre trial of groin hernia repair, supplemented by data from non-randomised operations performed during the trial. RESULTS - HEALTH TECHNOLOGY ASSESSMENT LITERATURE REVIEW: Of 4571 abstracts identified, 272 (6%) were later included in the study after review of the full paper. Some 51% of studies assessed a surgical minimal access technique and 95% were case series. The statistical method used most often (60%) was splitting the data into consecutive parts (such as halves or thirds), with only 14% attempting a more formal statistical analysis. The reporting of the studies was poor, with 31% giving no details of data collection methods. RESULTS - NON-HEALTH TECHNOLOGY ASSESSMENT LITERATURE SEARCH: Of 9431 abstracts assessed, 115 (1%) were deemed appropriate for further investigation and, of these, 18 were included in the study. All of the methods for complex data sets were identified in the non-clinical literature. These were discriminant analysis, two-stage estimation of learning rates, generalised estimating equations, multilevel models, latent curve models, time series models and stochastic parameter models. In addition, eight new shapes of learning curves were identified. RESULTS - TESTING OF STATISTICAL METHODS: No one particular shape of learning curve performed significantly better than another. The performance of 'operation time' as a proxy for learning differed between the three procedures. Multilevel modelling using the laparoscopic cholecystectomy data demonstrated and measured surgeon-specific and confounding effects. The inclusion of non-randomised cases, despite the possible limitations of the method, enhanced the interpretation of learning effects. CONCLUSIONS - HEALTH TECHNOLOGY ASSESSMENT LITERATURE REVIEW: The statistical methods used for assessing learning effects in health technology assessment have been crude and the reporting of studies poor. CONCLUSIONS - NON-HEALTH TECHNOLOGY ASSESSMENT LITERATURE SEARCH: A number of statistical methods for assessing learning effects were identified that had not hitherto been used in health technology assessment. There was a hierarchy of methods for the identification and measurement of learning, and the more sophisticated methods for both have had little if any use in health technology assessment. This demonstrated the value of considering fields outside clinical research when addressing methodological issues in health technology assessment. CONCLUSIONS - TESTING OF STATISTICAL METHODS: It has been demonstrated that the portfolio of techniques identified can enhance investigations of learning curve effects. (ABSTRACT TRUNCATED)
Cholecystectomy↗
[Evaluation of using statistical methods in selected national medical journals].
The paper covers the performed evaluation of frequency with which the statistical methods were applied in analyzed works having been published in six selected, national medical journals in the years 1988-1992. For analysis the following journals were chosen, namely: Klinika Oczna, Medycyna Pracy, Pediatria Polska, Polski Tygodnik Lekarski, Roczniki Państwowego Zakładu Higieny, Zdrowie Publiczne. Appropriate number of works up to the average in the remaining medical journals was randomly selected from respective volumes of Pol. Tyg. Lek. The studies did not include works wherein the statistical analysis was not implemented, which referred both to national and international publications. That exemption was also extended to review papers, casuistic ones, reviews of books, handbooks, monographies, reports from scientific congresses, as well as papers on historical topics. The number of works was defined in each volume. Next, analysis was performed to establish the mode of finding out a suitable sample in respective studies, differentiating two categories: random and target selections. Attention was also paid to the presence of control sample in the individual works. In the analysis attention was also focussed on the existence of sample characteristics, setting up three categories: complete, partial and lacking. In evaluating the analyzed works an effort was made to present the results of studies in tables and figures (Tab. 1, 3). Analysis was accomplished with regard to the rate of employing statistical methods in analyzed works in relevant volumes of six selected, national medical journals for the years 1988-1992, simultaneously determining the number of works, in which no statistical methods were used. Concurrently the frequency of applying the individual statistical methods was analyzed in the scrutinized works. Prominence was given to fundamental statistical methods in the field of descriptive statistics (measures of position, measures of dispersion) as well as most important methods of mathematical statistics such as parametric tests of significance, analysis of variance (in single and dual classifications). non-parametric tests of significance, correlation and regression. The works, in which use was made of either multiple correlation or multiple regression or else more complex methods of studying the relationship for two or more numbers of variables, were incorporated into the works whose statistical methods were constituted by correlation and regression as well as other methods, e.g. statistical methods being used in epidemiology (coefficients of incidence and morbidity, standardization of coefficients, survival tables) factor analysis conducted by Jacobi-Hotellng's method, taxonomic methods and others. On the basis of the performed studies it has been established that the frequency of employing statistical methods in the six selected national, medical journals in the years 1988-1992 was 61.1-66.0% of the analyzed works (Tab. 3), and they generally were almost similar to the frequency provided in English language medical journals. On a whole, no significant differences were disclosed in the frequency of applied statistical methods (Tab. 4) as well as in frequency of random tests (Tab. 3) in the analyzed works, appearing in the medical journals in respective years 1988-1992. The most frequently used statistical methods in analyzed works for 1988-1992 were the measures of position 44.2-55.6% and measures of dispersion 32.5-38.5% as well as parametric tests of significance 26.3-33.1% of the works analyzed (Tab. 4). For the purpose of increasing the frequency and reliability of the used statistical methods, the didactics should be widened in the field of biostatistics at medical studies and postgraduation training designed for physicians and scientific-didactic workers.
Case-Control Studies↗
Selecting ozone exposure statistics for determining crop yield loss from air pollutants.
Numerous ozone exposure statistics were calculated using hourly ozone data from crop yield loss experiments previously conducted for alfalfa, fresh market and processing tomatoes, cotton, and dry beans in an ambient ozone gradient near Los Angeles, California. Exposure statistics examined included peak (maximum daily hourly) and mean concentrations above specific threshold levels, and concentrations during specific time periods of the day. Peak and mean statistics weighted for ozone concentration and time period statistics weighted for hour of the day were also determined. Polynomial regression analysis was used to relate each of 163 ozone statistics to crop yield. Performance of the various statistics was rated by comparing residual mean square (RMS) values. The analyses demonstrated that no single statistic was best for all crop species. Ozone statistics with a threshold level performed well for most crops, but optimum threshold level was dependent upon crop species and varied with the particular statistics calculated. The data indicated that daily hours of exposure above a critical high-concentration threshold related well to crop yield for alfalfa, market tomatoes, and dry beans. The best statistic for cotton yield was an average of all daily peak ozone concentrations. Several different types of ozone statistics performed similarly for processing tomatoes. These analyses suggest that several ozone summary statistics should be examined in assessing the relationship of ambient ozone exposure to crop yield. Where no clear statistical preference is indicated among several statistics, those most biologically relevant should be selected.
Journal Article↗
A powerful and robust new linkage statistic for discordant sibling pairs.
Previously, Szatkiewicz and colleagues evaluated the performance of a wide variety of statistics for quantitative-trait-locus linkage, using discordant sibling pairs. They found that the most powerful statistics, in general, were a score statistic and a "composite statistic." However, whereas these two statistics have equal power under ideal conditions, each has limitations that reduce its power in certain circumstances. The score statistic depends on estimates of trait parameters and can lose a lot of power if those estimates are incorrect. The composite statistic is not sensitive to trait-parameter estimates but does depend on arbitrary weights that must be chosen on the basis of the ascertainment scheme. In this report, we elucidate the algebraic relationship between the score and composite statistics and then use that relationship to suggest a new statistic that combines the best properties of both. We call our new statistic the "robust discordant pair" (RDP) statistic. We report simulation studies to show that the RDP statistic does, indeed, have all of the strengths and none of the weaknesses of the score and composite statistics.
Computer Simulation↗
The teaching of health statistics: meeting the needs of a changing practice.
The findings of the study did support the faculty's hypotheses. First, only 50% of the hospitals retain responsibility for the statistical function. This finding was, however, affected by the size of the hospital, with 100% of the hospitals of 400 beds or less retaining such responsibility. Second, the findings showed that only 12.5% of the hospitals had entirely manual statistical systems. However, 50% of the hospitals did compile some statistics manually, including the daily hospital census and discharge service statistics. Finally, in looking at the UT Memphis statistics curriculum and those of 14 other medical record administration programs in the Southeast, the hypothesis that curricula did not mirror this changing practice was confirmed. Although 100% of the programs surveyed had students memorize statistical formulas, only 36% had students working with computers in the statistics course. Of the 14 programs, only 6 specifically covered QA of statistical data in the statistics course, and 9 did not deal at all with how to assess the adequacy of policies and procedures for gathering statistical information. As a result of these findings, UT Memphis has modified its statistical course to increase the emphasis on computerization, QA, and assessment of statistical policies and procedures. These changes will better prepare the UT Memphis graduate for the statistical responsibilities that they will face in the workplace.
Curriculum↗
A survey of statistical methodology used in Ethiopian health science research journals.
Results from many research efforts have been generated from the use of statistical methods. However, most researchers use data analysis as the only component of statistics to arrive at their results. There is a concern that this alone may not yield the appropriate result if it is not done with due understanding of and regard for study design, data acquisition techniques, choice of sample, and methods of statistical analysis. This paper attempts to document how much of these statistical methods are in use in Ethiopian health science research journals. All the original articles, in the two health science research journals--Ethiopian Medical Journal and The Ethiopian Journal of Health Development published between 1995 and 1999 were surveyed. A total of 232 papers were evaluated to see how far their authors have complied to these basic requirements as well as the statistical software used. The results of the survey demonstrate that in about 80% of the papers, the study design has been specified, 50% employed cross-sectional or survey designs, 14% provided detailed information on how sample size was determined and of this group 37% employed probabilistic selection methods. About 84% of the papers did not mention clearly what statistical methods they intended to employ to answer their research questions. Compared to others, attaching variability to a statistic using +/- SD or SE, t-statistics and P values were more frequently misused. Only 57% used computers to manage their data and do statistical analysis. EPI-INFO (a statistical software for Epidemiology) was used in 61% among the users of computer software. Considering the important roles of health science journals in guiding and updating good medical practice, low level and inappropriate use of statistical methodologies in the surveyed journals should give cause for concern. It is, therefore, recommended that a series of continuing education in statistics is done periodically to enhance the knowledge of health science researchers as well as editors and peer reviewers of health science journals to expand their background in statistical methods and acquaint them with new techniques.
Ethiopia↗
Nonparametric simulation-based statistics for detecting linkage in general pedigrees.
We present here four nonparametric statistics for linkage analysis that test whether pairs of affected relatives share marker alleles more often than expected. These statistics are based on simulating the null distribution of a given statistic conditional on the unaffecteds' marker genotypes. Each statistic uses a different measure of marker sharing: the SimAPM statistic uses the simulation-based affected-pedigree-member measure based on identity-by-state (IBS) sharing. The SimKIN (kinship) measure is 1.0 for identity-by-descent (IBD) sharing, 0.0 for no IBD status sharing, and the kinship coefficient when the IBD status is ambiguous. The simulation-based IBD (SimIBD) statistic uses a recursive algorithm to determine the probability of two affecteds sharing a specific allele IBD. The SimISO statistic is identical to SimIBD, except that it also measures marker similarity between unaffected pairs. We evaluated our statistics on data simulated under different two-locus disease models, comparing our results to those obtained with several other nonparametric statistics. Use of IBD information produces dramatic increases in power over the SimAPM method, which uses only IBS information. The power of our best statistic in most cases meets or exceeds the power of the other nonparametric statistics. Furthermore, our statistics perform comparisons between all affected relative pairs within general pedigrees and are not restricted to sib pairs or nuclear families.
Algorithms↗
Research design and statistical methods in Chinese medical journals.
CONTEXT: Study design and statistical analyses have improved in journals published in Western countries, but the type of research designs and statistical methods used in medical journals outside Western countries has not been assessed. OBJECTIVES: To determine the frequency of research designs and statistical techniques used in Chinese medical journals, types of statistical errors present, and trends over a 10-year period. DESIGN: Evaluation of all original articles published in 5 leading journals in 1985 (N = 640) and in 1995 (N = 954). MAIN OUTCOME MEASURES: Research designs and statistical methods. RESULTS: Compared with 1985, significant improvement was seen in 1995; the percentages of original articles reporting clinical trials, prospective studies, or basic science research increased from 18% to 31% (P<.001), the proportion of papers using statistical tests increased from 40% to 60% (P<.001), more sophisticated statistical methods were used, and of those articles using statistics, the proportion using appropriate methods increased from 22% to 46% (P<.001). In both years, the most commonly used statistical methods were t tests and contingency tables. The most common errors were presentation of P values without specifying the test used, use of multiple t tests instead of analysis of variance, and use of unpaired t tests when paired tests were required. CONCLUSIONS: The use of statistical methods in Chinese medical journals research is improving, and by 1995, the frequency of using statistical methods in published articles was similar to the results determined in previous studies of journals. However, the lack of or inappropriate use of statistics remains a serious problem.
China↗
Statistical methods in the surgical literature.
BACKGROUND: It is important that clinicians understand statistical methods to incorporate statistics into their own research and to correctly translate published literature into improved patient care. The purpose of this study was to identify frequency and appropriate use of statistical methods in clinical surgical publications during the past 18 years. STUDY DESIGN: The study included randomly selected issues from odd-numbered years of Annals of Surgery (Annals) and Archives of Surgery (Archives) between 1985 and 2003, and issues in 2003 from Journal of the American College of Surgeons (JACS), Journal of Surgical Research (JSR), and Surgery. We identified all statistical procedures reported in each article, examined correctness of methods, and reported trends in publication of statistical methods over time. RESULTS: The proportion of publications incorporating statistics has increased over time. Declining trends were seen in the proportion of articles with no statistics (p < 0.0001). Approximately 35% of articles in 1985 did not use statistics compared with < 10% in 2003. Nonparametric tests increased (p < 0.0001) during the study period. In Archives of Surgery, nonparametric tests increased from 0% in 1985 to 33% in 2003, and in Annals of Surgery, from 12% in 1985 to 49% in 2003. Twenty-seven percent of studies included incorrect selection or reporting of statistical methods. CONCLUSIONS: Overall, the statistical complexity of research in clinical surgery journals is increasing. It is important that clinicians reading this literature have sufficient knowledge of statistical methods to facilitate interpretation of increasingly sophisticated statistical analyses.
General Surgery↗