Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “statistical inference”

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 685 records · Page 38Linked to original sources
PubMed2005

Bioequivalence of two brands of citalopram 40 mg tablets after single oral administration to healthy volunteers.

A randomized, two-way, crossover, bioequivalence study was conducted in 26 fasting, healthy, male volunteers to compare two brands of citalopram 40 mg tablets, Citol (Abdi Ibrahim Ilaç San. ve Tic A.S., Istanbul, Turkey) as a test and Cipramil (H. Lundbeck A/S, Copenhagen, Denmark) as a reference product. One tablet of either formulation was administered with low-carbonate water after 10 h of overnight fasting. After dosing, serial blood samples were collected during a period of 24 hours. Plasma samples were analysed for citalopram by a validated HPLC method. The pharmacokinetic parameters AUC0-24, AUC(0-alpha), Cmax, Tmax, K(el), T(1/2), and CL were determined from plasma concentration-time profiles for both formulations and were compared statistically to evaluate bioequivalence between the two brands of citalopram, using the statistical modules recommended by FDA. The analysis of variance (ANOVA) did not show any significant difference between the two formulations and 90% confidence intervals (CI) fell within the acceptable range for bioequivalence. Based on these statistical inferences it was concluded that the two brands exhibited comparable pharmacokinetics profiles and that Abdi Ibrahim's Citol is equivalent to Cipramil of H. Lundbeck, Copenhagen, Denmark.

Administration, Oral↗
PubMed1991

Class A results in Menière's disease.

A series of 101 patients with Menière's disease treated surgically with an intramastoid endolymphatic shunt operation and with a minimum follow-up of 2 years is reported. The results included success in 76 percent and failure in 22 percent. Thirteen of these patients maintained a hearing level of 15 dB or better, while 11 patients experienced hearing gains of 25 dB or better. These results are compared to other reports, and the difficulty of making such comparisons is illustrated. A derived "efficacy ratio" is defined, and its use is suggested to aid in making statistical inference. The conclusion is reached that more knowledge is required about the natural history of Menière's disease so that appropriate staging may be carried out on these study populations.

Endolymphatic Shunt↗
PubMedDate not supplied

Computerized expert system for the diagnosis of pulp-related pain.

A major problem in the correct diagnosis of pulpal pain is that the associated clinical signs do not predictably correlate with the underlying pathological process. Using conditional probabilities of various pulp conditions from published data, Bayesian Statistical Inference provides the means for deriving a composite probability of the presence of a disease from a multiple set of symptoms. A computer program that can infer a diagnosis for pulpal pain from any combination of 17 clinical symptoms has been developed. From the data, the program provides the computed relative probabilities of a healthy pulp, a saveable pulp, an unsaveable pulp, and a necrotic pulp being present.

Bayes Theorem↗
PubMed1990

[Rheumatology research. An analysis of the situation in Spain].

To analyze the present status of rheumatological research in Spain, three indicators have been evaluated: use of the different study designs, use of statistical techniques and relevance of the studies. This last point was determined by the comparison of the research areas that had been reported during one year and the lines of rheumatological research established as priorities by 10 investigators from the 10 units of clinical care with a higher scientific output during the period 1985-1987. A revision has been made of the 67 articles from Spain out of the 1106 reported in seven of the journals with a better national and international impact index: Scandinavian Journal of Rheumatology, Arthritis and Rheumatism, Journal of Rheumatology, Annals of the Rheumatic Diseases, British Journal of Rheumatology, Clinical Experimental Rheumatology and Revue du Rhumatisme, in addition to Revista Española de Reumatología. It was found that few studies with analytical design are being published and that complex statistical analysis techniques are not being used; therefore, most articles do not have a possible statistical inference and lack value as predictors. The reports of cases and clinical series in the Letters to the Editor section are most common, particularly in the Spanish journal. Most studies are related with clinical research, although the consulted investigators consider that basic and epidemiological research have a higher priority. It is concluded that the Spanish contribution to the international rheumatological literature is still small and has a mainly descriptive character; thus, it is necessary to emphasize the use of analytical designs to improve the competitiveness. The type of studies to be carried out in the future, according to the priorities established by the investigators themselves, should change if their wishes are to be fulfilled.

Humans↗
PubMed1990

The Cox proportional hazards model with change point: an epidemiologic application.

In this paper, we develop the Cox proportional hazards model with special structured time-dependent covariates in the context of prospective epidemiologic studies. Our model possesses the following two features: (i) different relative risk parameters are allowed for early versus late onset of the disease of interest; (ii) an additional parameter is introduced so that specification is not required for the time (age) at which a change of the magnitude of the relative risks takes place, the so-called change point. Some difficulties with statistical inference for the proposed model are briefly discussed, and the large-sample distribution of a test for no change point is derived. As an illustration, we apply the model to a set of data gathered on a group of white male medical students of The Johns Hopkins Medical School enrolled between 1948 and 1964. We examine the hypothesis that the effect of reactivity to the cold pressor test may vary with early versus late onset of hypertension.

Biometry↗
PubMed1990

Some comments on teaching biostatistics in medical and health sciences.

The role of statistical methods is now well recognized in health sciences since these disciplines are concerned with the study of communities or populations where the principles of sampling and statistical inference are clearly applicable. However, many medical and health sciences teachers and students have been slower to perceive the need for knowledge of biostatistical methods, even though all aspects of medical diagnosis and prognosis are governed by the laws of probability. Some of them are still skeptical about the value and importance of biostatistical principles to their fields and raise questions about the meaning, content, and nature of biostatistics and relevance of its teaching to health sciences disciplines. The purpose of this essay is to address some of these issues with the hope to invoke comments and responses from other biostatistics instructors who have encountered similar predicaments in their teaching and consulting roles to health sciences students and professionals.

Biometry↗
PubMedDate not supplied

[Possibilities and limitations of the predictive risk estimates and epidemiological studies following the Chernobyl incident].

The disastrous accident at the nuclear power station at the Chernobyl on 1986 (April 26) has brought attention to the estimation of radiation health effects and many "experts" were attending to the evaluation on oncogenic mortality increase among the Italian population in the next future. On the contrary at that time too few peoples were worried about the possibility of detecting such an increase. Discussion of this topic is notoriously fraught with difficulties arising from differences of opinion how to estimate low-dose risk in humans without data from direct observation. One opinion is to extrapolate from the data points obtained at relatively high doses toward zero dose (zero extrapolation theory). This permit estimates of risk to be made but, in the final analysis, no data from humans exist that show that low-level radiation exposures produce measurable biologic effects. For that this theory is more useful in radio-protection and medico-legal subjects. It is easy on a statistical basis to prove the impossibility to establish an increase in human cancer after low doses of ionizing radiation such as those received environmentally after the Chernobyl's accident. In this condition to observe the numbers of radiation-induced cancer deaths that far exceed the "natural" incidence would require a follow-up a sample more and more greater than the italian population herself. Indeed the statistical power of a hypothetical follow-up study at a suitable confidence level would require a sample size higher than a milliard of persons for the detection of an increase of a generic cancer mortality and higher then seven hundred of millions for the detection of an increase of the specific thyroid cancer mortality. In more detail the following figures for the parameters needed to curring out the evaluation have been used: medium dose equivalent to the thyroid, 2.03 mSv; medium effective dose equivalent up to december '87, 0.6 mSv; thyroid cancer mortality in the italian population, 0.94 10(-5) y-1; total cancer mortality in the italian population, 22.2 10(-2) y-1; risk factor per unit dose equivalent in thyroid, 0.5 10(-6) mSv-1; risk factor per unit effective dose equivalent, 2.0 10(-5) mSv-1. Applying the foregoing values in statistical inference methods it could be achieved that 7.5 18(8) and 1.25 10(9) persons must be followed-up in the next 30 years to detect a significant increase over the "natural" cancer mortality for thyroid and "total body" radioinduced cancers respectively.(ABSTRACT TRUNCATED AT 400 WORDS)

Accidents↗
PubMed1986

Peripheral blood T lymphocyte subpopulations determined by monoclonal antibodies in active rheumatoid arthritis.

The introduction of an automated flow cytofluorograph has facilitated the detection of T lymphocyte subsets because it enables a larger number of cells to be analyzed. Many authors have reported a decrease of cytotoxic/suppressor T lymphocytes in rheumatoid arthritis (RA), in contrast to the results of other workers. We believe that the discrepancy between the various studies is due to the fact that different methods and different criteria for patient selection were used. Our study comprised a larger number of patients which makes the results suitable for statistical inference. Disease activity is clearly defined and all drugs that could alter the results were excluded. The use of a flow cytometer enhances the reliability of T lymphocyte subset determination by monoclonal antibodies (OKT series). Our study confirms the reports, which suggested that the number of suppressor/cytotoxic T lymphocytes (OKT8+ cells/mm3) is decreased in patients with active RA, resulting in a high T helper/inducer lymphocyte/T suppressor/cytotoxic lymphocyte ratio (OKT4+:OKT8+). This immune balance represents an interesting feature of the disease, since several active antirheumatic drugs share a common immunomodulatory action, which normalizes the OKT4+:OKT8+ ratio. Finally, we found a good correlation between the OKT4+ cells and OKT8+ cells in the normal control population. This observation enabled us to isolate a subgroup of active patients with RA not responding to slow acting antirheumatic drugs and characterized by a low OKT4+:OKT8+ ratio.

Adult↗
PubMed1986

Use and misuse of epidemiologic data in the courtroom: defining the limits of inferential and particularistic evidence in mass tort litigation.

Medical epidemiology is the cornerstone for understanding the safety and efficacy of drugs and medical devices. Epidemiologic principles provide a statistical basis for determining correlations, and ultimately mathematical causation, between two series of events. Medical epidemiologic evidence and statistical inferences are useful and are now routinely accepted in the courtroom. The complex distribution systems that provide fungible goods throughout the country often preclude the identification of the specific source of an allegedly defective product. An expansion of the principles established in Summers v. Tice and Ybarra v. Spangard provide a logical and rational means for the courts to address products liability issues in cases involving multiple and unnamed defendants. This Article discusses the impact of epidemiology on the judicial process, both in the nature of judicial decision-making and in the nature of the law itself. Part III B discusses the "weak" and "strong" versions of the traditional preponderance of the evidence rule, as recast by recent products liability litigation. The remainder of the Article defines the useful and appropriate scope of epidemiologic evidence, concluding that "intentless" epidemiologic evidence alone cannot support an award of punitive damages.

2,4,5-Trichlorophenoxyacetic Acid↗
PubMed1986

Statistical quality control methods in standardization of BCG vaccines.

The quality control of the BCG vaccine lacks the requirements of exact statistical methods, although the principles of bioassay have been required in the control of other biological products for a long time. For a remedy and to eliminate controversies in the interpretation of the laboratory results of BCG products valid statistical control is needed. In the four major laboratory test systems generally used to control: the viability of the BCG products, the residual virulence of the BCG strains, the allergenic potency and 4) the immunogenic potency of the BCG vaccines the following statistical methods should be required: estimations of the unbiased point and 95% confidence interval of the viable units 10(6)/ml, the mean residual virulence, the mean allergizing and protective capacity in appropriate models, the expressions of the relative potency of the in vitro viability and the in vivo test systems to a Reference Preparation, hypothesis tests using the unbiased estimates for the comparison of the estimated parameters of the controlled vaccines (Anova) and testing the stability of the viability (Regranal), multiple comparison with the ranked estimates to diminish Type I error in rejected null hypotheses, test of concordance between the sets of ranked parameters of the different test systems. Valid models will promote the standardization of the BCG vaccine, will ensure the objective statistical inference of the control results, the interpretation of the immunogenicity and reactogenicity of the BCG vaccines and the understanding of the epidemiological processes in tuberculosis.

Animals↗
PubMed1986

Quantitative methods in the tuberculosis epidemiology and in the evaluation of BCG vaccination programs.

Controversies concerning the protective efficacy of the BCG vaccination result mostly from the fact that quantitative methods have not been used in the evaluation of the BCG programs. Therefore, to eliminate the current controversy an unconditional requirement is to apply valid biostatistical models to analyse the results of the BCG programs. In order to achieve objective statistical inferences and epidemiological interpretations the following conditions should be fulfilled: data for evaluation have to be taken from epidemiological trials exempt from sampling error, since the morbidity rates are not normally distributed an appropriate normalizing transformation is needed for point and confidence interval estimations, only unbiased point estimates (dependent variables) could be used in valid models for hypothesis tests, in cases of rejected null hypothesis the ranked estimates of the compared groups must be evaluated in a multiple comparison model in order to diminish the Type I error in the decision. The following quantitative methods are presented to evaluate the effectiveness of BCG vaccination in Hungary: linear regression analysis, stepwise regression analysis and log-linear analysis.

Adolescent↗
PubMedDate not supplied

Morphologic and functional features of cyclic alternating pattern (CAP) sequences in normal NREM sleep.

The Cyclic Alternating Pattern (CAP) is an EEG biphasic periodic activity of normal NREM sleep, generally organized in sequences of two or more cycles and correlated to long-lasting fluctuations in the level of arousal. The remaining NREM sleep is characterized by prolonged stationary activities or non-CAP (NCAP). Criteria for the detection of morphology and reactivity of CAP and NCAP segments are defined in the sleep recordings of 10 healthy young adults. Standard sleep variables and CAP rates, referred to the ratio of CAP time to Total NREM sleep time (CAPR/NREM) and to each NREM sleep stage, were scored for statistical inferences and correlations. CAPR/NREM may be considered an additional non-redundant sleep variable which determines the measure of arousal instability during NREM sleep. The relationship between CAP and the biphasic behavior of microarousals is discussed.

Acoustic Stimulation↗
PubMed1984

Approximate case influence for the proportional hazards regression model with censored data.

A method is presented for approximating the influence of individual cases upon regression coefficient estimates obtained from the Cox proportional hazards model. Observations can thus be identified which may greatly influence statistical inferences regarding the effects of prognostic factors upon survival time. An example from a cancer clinical trial is given.

Biometry↗
PubMedDate not supplied

[Demonstration of bencyclane effects in a clinico-pharmacologic vigilance model].

The effects of 8 weeks' treatment with 2x 200 mg Bencyclan daily were investigated versus placebo in 50 patients. A clinical diagnosis had revealed the onset of organic mental disorders in all patients who, after fulfilling neurophysiologically and neuropsychologically defined screening criteria, were admitted to the trial. The patients were randomly allocated to two parallel groups of 25 for this controlled, double-blind study. Patients had to show 50% subvigil phases in a 15 min resting-EEG recording. We define such behaviour as a neurophysiological disturbance of vigilance. Scores in the Benton test were to be 2 points below the expected value or the NAF-score was to be above a standard value attained in an old people's home (greater than 14). On a neurophysiological level, inference statistical tests indicated effects in line with the hypothesis: A statistically significant difference in favour of Bencyclan was apparent after 8 weeks' treatment when compared with placebo. The Vigilance Index (VI) was higher under Bencyclan than after placebo. We regard the VI as a plausible indicator of vigilance-dependent disturbances in the elderly. On a behavioral level, Bencyclan was tendentially superior with respect to age-dependent complaints and performance-related activatedness, whereby one could not exclude the possibility that there were chance findings. The clear proof of efficacy on the neurophysiological level and the unclear effectiveness on the behavioral level can be attributed to the fact that time- and placebo effects were less marked on the former level. The intensive supervision of the patients led to considerable placebo effects on the behavioral level. Two groups of 25 patients are therefore not sufficient to be able to detect drug effects within a treatment period of only 8 weeks.

Aged↗
PubMed1983

A unified model for complex segregation analysis.

Various methods have been proposed for statistical inference of major genes by segregation analysis of human familial data. An attempt is made to resolve some divergences that have occurred in this context by the consideration of a unified model, with some practical applications.

Alleles↗
PubMed1982

[Hemispheric lateralization of EEG in early infantile autism].

Autoregressive (AR) analysis and component analysis were applied to EEGs during sleep stage 2 of normal and autistic children, since AR-coefficients of EEG characterize the autocovariance and power spectral density of EEG. In addition, as the Mahalanobis' generalized distance of the average AR coefficient vector of an EEG group at one side from that at the opposite side follows a F-distribution, the statistical inferences were carried out, whether or not there is hemispheric specialization in EEG in 21 right handed autistic children and whether there is difference between AR-spectral pattern of EEG in the autistic children and that of 28 normal children of the same age range or not. The results obtained were as follows: 1) The significant hemispheric lateralization in EEG at the significant level of 0.01 was observed in normals, whereas none of the lateralization was confirmed in autistic children. 2) The random click stimulation caused a cumulative effect on the hemispheric lateralization in the normals, but not in the autistic children. On the contrary, the effects of random flash light stimulation on hemispheric asymmetry were verified in the autistic children as well as in the normals. 3) The EEG components of autistic children were significantly different from the normal ones in frequency range of alpha-2 component wave at bilateral frontal (F1 and F2) regions and the activity of these component waves were higher in the autistic children than in the normal ones over 5 years-old. 4) These findings suggest that autistic children are not only less responsive to external stimuli, but also neurophysiologically less active in the internal regulatory mechanism.

Acoustic Stimulation↗
PubMed1995

Decision analysis: theory and application to medicine.

The formal methodology of clinical decision analysis has its origins in probability theory, statistical inference, game theory, and economics. Its application to the arena of problems in medicine and public policy should have a vast impact on the way medicine is practiced in the twenty-first century and beyond. The clinical decision analyses published in journals are based on particular assumptions; if these assumptions do not hold true, the results must be regarded as suspect. However, if the assumptions do hold true or if they reasonably approximate reality, the analyses and results of decision analyses must be appreciated and understood. Their interpretation may yield improved health for patients, monetary savings, or an explicit understanding of the uncertainties that exist and the tradeoffs that must be considered in decision making in primary care.

Animals↗
PubMed1993

Interactive multivariate analysis of bisindolylmaleimides as potent protein kinase C antagonists.

The isoenzyme protein kinase C (PKC) inhibitory activity of substituted bisindolylmaleimides depends on the molecular weight, the total charge, and dipole moments. The validity of the resulting QSAR equation was investigated by interactive diagnostic statistics and multivariate simultaneous statistical inference. Molecular mechanics and dynamics can be used to study possible reasons of flagged observations (high-leverage points, influential data, outliers) of QSAR systems.

Maleimides↗