Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Statistical Regression”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

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

At least 1,747 records · Page 97Linked to original sources
PubMed1997

Methods of correcting for multiple testing: operating characteristics.

We examine the operating characteristics of 17 methods for correcting p-values for multiple testing on synthetic data with known statistical properties. These methods are derived p-values only and not the raw data. With the test cases, we systematically varied the number of p-values, the proportion of false null hypotheses, the probability that a false null hypothesis would result in a p-value less than 5 per cent and the degree of correlation between p-values. We examined the effect of each of these factors on family-wise and false negative error rates and compared the false negative error rates of methods with an acceptable family-wise error. Only four methods were not bettered in this comparison. Unfortunately, however, a uniformly best method of those examined does not exist. A suggested strategy for examining corrections uses a succession of methods that are increasingly lax in family-wise error. A computer program for these corrections is available.

Clinical Trials as Topic↗
PubMed1997

Discriminant analysis using the unweighted sum of binary variables: a comparison of model selection methods.

Many clinical decision-making rules are equivalent to linear discriminant functions that involve the unweighted sum of binary variables (SBV). We briefly consider the geometry of this restriction and then propose a number of methods for forward stepwise selection of SBV models. Using a simulation study, we compare the performance of these methods under a wide range of plausible conditions and show that no single method is uniformly superior for selecting models of a fixed size. Factors of general importance in relative method performance are the ratio of sample size to the number of candidate variables and the class-conditional moment structure of the data. We conclude by offering some practical strategies for SBV model construction.

Computer Simulation↗
PubMed2001

Psychosocial and health status variables independently predict health care seeking in fibromyalgia.

OBJECTIVE: To determine whether variables derived from the self-regulatory model of health and illness behavior accurately predict status as a patient or nonpatient with fibromyalgia (FM). METHODS: Subjects were 79 patients who met American College of Rheumatology (ACR) criteria for FM and 39 community residents who met ACR criteria for FM but had not sought medical care for their symptoms (nonpatients). Subjects were administered 14 measures that produced 6 domains of variables: background demographics and pain duration; psychiatric morbidity; and personality, environmental, cognitive, and health status factors. These domains were entered in 4 different hierarchical logistic regression analyses to predict status as patient or nonpatient. RESULTS: The full regression model was statistically significant (P < 0.0001) and correctly identified 90.7% of the subjects with a sensitivity of 92.4% and a specificity of 87.2%. The best individual predictors of group status were self-reports of self-efficacy, negative affect, recent stressful events, and perceived pain. Relative to nonpatients, patients reported higher levels of negative affect and perceived pain and a greater number of recent stressful experiences, as well as lower levels of self-efficacy. CONCLUSION: Consistent with the self-regulatory model of health and illness behavior, psychosocial and health status variables predict health care-seeking behavior in persons with FM independently of background demographics and psychiatric morbidity. These variables may influence the severity of symptoms experienced by persons with this disorder as well as their health care-seeking behavior, but they are not necessary to produce abnormal pain sensitivity in FM.

Activities of Daily Living↗
PubMed2003

Predicting difficult orotracheal intubation in pharyngo-laryngeal disease: preliminary results of a composite index.

PURPOSE: Prediction of difficulty in orotracheal intubation (DI) in patients undergoing laryngeal microsurgery should help reduce the morbidity-mortality associated with this clinical situation. To establish a simple score to predict this difficulty, we studied 11 variables and their association with DI in these patients. METHODS: The study included 181 patients. The variables evaluated were: Mallampati grade, thyromental distance, mouth opening, temporomandibular joint movement, tooth morphology, maxillary deficiency, head and neck movement, receding mandible, body mass index, and clinical symptoms of laryngeal and supraglottic disease. To establish the score, regression coefficients of the statistically significant variables were used on adjusted logistic regression analysis. RESULTS: DI was present in 50 patients (28%) and orotracheal intubation was impossible in four (2%). Except for obesity, all the variables evaluated were predictive of DI. A simple predictive test was established based on logistic regression analysis including all the variables except temporomandibular joint movement. To determine the optimum cut-off for the new test, a receiver operating characteristic curve analysis was applied. A score > or = 5 in the proposed test provided a sensitivity of 94% and a specificity of 76%. CONCLUSION: The index we describe is aimed at predicting DI in a very specific population with a high risk of this complication. This index uses a series of variables which may be measured easily during the preoperative period and provides an excellent predictive capacity with a high sensitivity and specificity when the index is > or = 5.

Female↗
PubMed2006

Ethnopharmacological evaluation of the informant consensus model on anti-tuberculosis claims among the Manus.

Ethnobotanists often utilize predictive models to analyze the potential of indigenously used medicinal plants. The most common of these prognostic models is the informant consensus model. This study evaluates use of this model through the analytical ethnopharmacology of Manus Province, Papua New Guinea (PNG). The informant consensus model enables researchers to prioritize plants for pharmacognostic evaluation, based on the relative frequency of plants cited in anthropological interviews. Fieldwork on Manus Island, PNG, led to the identification of 43 species of plants used in traditional medicine for persistent respiratory symptoms. Plants were collected, dried, micro-extracted using a new technique generated in our laboratory, and evaluated in vitro against Mycobacterium tuberculosis. The results, in the form of IC(50) values and modified selectivity indices (SI), were compared to the results of the anthropological models of informant consensus, and statistically compared through linear regression and t-tests. Results were not statistically significant (alpha=0.1), leading to the conclusions that the informant consensus assumptions were inaccurate in predicting anti-mycobacterial activity among the Manus for anti-TB claims.

Anti-Bacterial Agents↗
PubMed2002

Effects of road geometry and traffic volumes on rural roadway accident rates.

This paper revisits the question of the relationship between rural road geometric characteristics, accident rates and their prediction, using a rigorous non-parametric statistical methodology known as hierarchical tree-based regression. The goal of this paper is twofold: first, it develops a methodology that quantitatively assesses the effects of various highway geometric characteristics on accident rates and, second, it provides a straightforward, yet fundamentally and mathematically sound way of predicting accident rates on rural roads. The results show that although the importance of isolated variables differs between two-lane and multilane roads, 'geometric design' variables and pavement condition' variables are the two most important factors affecting accident rates. Further, the methodology used in this paper allows for the explicit prediction of accident rates for given highway sections, as soon as the profile of a road section is given.

Accidents, Traffic↗
PubMed1999

The role of computed tomography scans in evaluating sinus disease in pediatric patients.

OBJECTIVES: To determine the incidence and severity of sinus abnormalities in children undergoing computed tomography (CT) of the sinuses for suspected chronic sinusitis. To compare these findings with abnormalities noted on random CT scans. METHODS: Sixty CT scans, performed for evaluation of sinus disease in symptomatic children aged 2-12, were compared with 50 CT scans of children aged 2-12 of the orbits or sinuses obtained for indications other than sinusitis. A staging system was applied to assess the severity of abnormalities. RESULTS: Mucoperiosteal thickening was present in 60% of symptomatic and 46%, of random CT scans (logistic regression, P = 0.144). Children aged 2-4 and 9-12 had an increased prevalence of abnormalities in both groups, although these findings were not statistically significant (logistic regression, P = 0.817). Early stage sinus disease was present in the majority of random (96%) and symptomatic (85%) children. CONCLUSIONS: There is a high incidence of mucoperiosteal thickening in the paranasal sinuses of children. CT scans of the sinuses should be obtained from children who are being considered for sinus surgery after failing the appropriate medical therapy. Decisions regarding the need for sinus surgery should not be solely based on imaging abnormalities.

Child↗
PubMed2006

Effect on adult height of pubertal growth hormone retesting and withdrawal of therapy in patients with previously diagnosed growth hormone deficiency.

CONTEXT: GH replacement therapy in GH-deficient (GHD) patients is usually continued until adult height despite the fact that most of these subjects display a normal secretion when retested at the end of growth. Puberty is the most likely time for normalization of GH secretion. OBJECTIVES: The objectives of this study are to establish the characteristics and the percentage of the subjects with isolated GHD who normalized secretion at puberty and to compare their statural outcomes with those of the subjects with persistent deficiency treated also after retesting. DESIGN AND SETTING: This was a prospective, nonrandomized, open-label study conducted in a university research hospital. PATIENTS AND INTERVENTION: Sixty-nine subjects (40 male, 29 female) with a diagnosis before puberty of isolated GHD by means of arginine and l-dopa tests were reevaluated with the same tests after at least 2 yr of therapy and after puberty onset. If GH peak at retesting was more than 10 microg/liter, therapy was withdrawn. MAIN OUTCOME MEASURES: Percentage and characteristics of normalized subjects at retesting, outcome of treatment in the subjects treated or untreated to adult height, and factors predictive of growth outcome were measured. RESULTS: At retesting, 44 subjects (63.7%) confirmed a GH peak less than 10 microg/liter (24 of 40 male and 20 of 29 female). Apart from a less delayed bone age at diagnosis in females, the subjects with confirmed GHD were not different at diagnosis from the other group for height deficit at diagnosis, first year growth response to GH, age and height at puberty onset, height, and IGF-I at retesting. Mean adult height was 165.1 +/- 4.5 cm in the male group treated until adult height vs. 164.0 +/- 3.4 cm in the group who suspended therapy at retesting. Mean adult height was 153.2 +/- 4.1 cm in the female group treated until adult height vs. 152.9 +/- 5.2 cm in the group that suspended therapy at retesting. As regards the parameters expressing the final outcome, the only difference was found in the mean increment adult height-target height sd score in favor of the male group treated until adult height. In both sexes, therapy duration and GH levels at diagnosis and at retesting were unrelated to adult height parameters and to height increments during the period of observation. CONCLUSIONS: One third of our GHD subjects diagnosed before puberty presented a normal secretion at puberty. The withdrawal of GH therapy in these subjects after retesting was not associated with a catch down growth, and they obtained an adult height similar to those obtained by the GHD subjects treated until adult height. It seems convenient, in subjects with nonsevere GHD, to retest GH secretion at midpuberty and to withdraw treatment for the subjects that are no longer deficient.

Adolescent↗
PubMed2004

Statistics in the pharmacy literature.

BACKGROUND: Research in statistical methods is essential for maintenance of high quality of the published literature. OBJECTIVE: To update previous reports of the types and frequencies of statistical terms and procedures in research studies of selected professional pharmacy journals. METHODS: We obtained all research articles published in 2001 in 6 journals: American Journal of Health-System Pharmacy, The Annals of Pharmacotherapy, Canadian Journal of Hospital Pharmacy, Formulary, Hospital Pharmacy, and Journal of the American Pharmaceutical Association. Two independent reviewers identified and recorded descriptive and inferential statistical terms/procedures found in the methods, results, and discussion sections of each article. Results were determined by tallying the total number of times, as well as the percentage, that each statistical term or procedure appeared in the articles. RESULTS: One hundred forty-four articles were included. Ninety-eight percent employed descriptive statistics; of these, 28% used only descriptive statistics. The most common descriptive statistical terms were percentage (90%), mean (74%), standard deviation (58%), and range (46%). Sixty-nine percent of the articles used inferential statistics, the most frequent being chi(2) (33%), Student's t-test (26%), Pearson's correlation coefficient r (18%), ANOVA (14%), and logistic regression (11%). CONCLUSIONS: Statistical terms and procedures were found in nearly all of the research articles published in pharmacy journals. Thus, pharmacy education should aim to provide current and future pharmacists with an understanding of the common statistical terms and procedures identified to facilitate the appropriate appraisal and consequential utilization of the information available in research articles.

Education, Pharmacy↗
PubMedDate not supplied

Corneal endothelial damage after neodymium:YAG laser iridotomy.

BACKGROUND AND OBJECTIVE: This study examined the long-term effect of neodynium:YAG (Nd:YAG) laser iridotomy on the corneal endothelium. PATIENTS AND METHODS: A prospective study was designed. Patients with narrow and occludable angles or fellow eye of acute angle closure glaucoma attack were treated with Nd:YAG laser iridotomy. For one year, 31 eyes of 21 patients underwent complete follow-up. Corneal endothelial specular microscopy was performed before and after laser iridotomy at 1, 3, 6, and 12 months. RESULTS: The decrease of endothelial cell density after YAG laser iridotomy was statistically significant at 1 month (P = 0.036), 6 months (P = 0.004), and 12 months (P = 0.000), respectively. The decrease was not statistically significant at 3 months (P = 0.467). Linear regression analysis indicated no statistical correlation between the percentage change in endothelial cell density and the total energy used during the treatment (1 month: P=0.08, 3 months: P= 0.3, 6 months: P=0.9, 12 months: P=0.2). CONCLUSION: This study demonstrated significant endothelial cell loss in the 1-year follow-up. The result suggested that Nd:YAG laser iridotomy may pose a long-term hazard to the corneal endothelium.

Adult↗
PubMed2000

[The predication equation of the permanent canine and premolar crown].

OBJECTIVE: It is important to predict mesiodistal crown diameter of the permanent canines and premolars in the mixed dentition, which helps to determine malocclusion and its further development. But there is still no article about this equation of Chinese used to predict mesiodistal crown diameter of cuspids and premolars. This research is to measure mesiodistal crown diameters, to study relationship between lower incisors and canines and premolars, and to provide the most accurate and easy basis in orthodontic diagnosis and treatment for Chinese people. METHODS: 140 Chinese students (60 male, 80 female) who were from eighteen to twenty-five years old without orthodontic history Chengdu were randomly collected in. All teeth must be free of visible malform, aberration, and fully erupted. Impressions were taken, and measurements were made, parallel to the occlusal plane at the greatest mesiodistal width of each tooth on dental cast with electronic caliper (0.01 mm). Each tooth was measured twice and the average of the two readings was taken as the dimension for that individual tooth. Mesiodistal crown diameters were statistically analyzed with t-test and linear regression, using the SAS (Statistic Analyses System) computer package. RESULTS: It was found that mesiodistal crown diameters had significantly statistical difference in gender. Hence the mesiodistal crown diameters were assessed and investigated separately for the male and the female. It was also found that the total mesiodistal crown diameter of lower incisors was closely correlated with the total mesiodistal crown diameter of canines and premolars. The correlation coefficients between the independent variable and dependent variables were 0.599 for upper canines and premolars and 0.654 for lower canines and premolars in the males and 0.586 and 0.635, respectively, in the females. The P were found to be less than 0.001 for both arches. The correlation coefficients had significantly statistical difference. CONCLUSION: The equations predicting overall mesiodistal crown diameters of upper and lower unerupted canines and premolars for male and female Chinese people were established respectively: Maxillary (y = 10.87 + 0.51 x (male)), y = 11.88 + 0.46 x (female), Mandibular (y = 10.36 + 0.50 x (male), y = 10.03 + 0.49 x (female)), y (total mesiodistal crown diameter of canines and premolars), and x (total mesiodistal crown diameter of lower incisors).

Adolescent↗
PubMedDate not supplied

Patterns of daily allocation of sleep periods: a case study in an Amazonian riverine community.

Few works already carried out have examined the relative role of genetic and external factors on the determination of the rhythmicity of the human sleep/wake cycle. In order to make a preliminary approach in this field, we investigated the diversity of patterns of allocation of sleep periods among 29 families living at the Combu Island, a socioculturally very homogeneous human group of the Brazilian Amazon. The individuals were interviewed through a questionnaire designed by Horne and Ostberg (1976), with the language of the questions adjusted to the way-of-life of the riverine people. A large predominance of the morning type was observed (95.35%), what constitutes a strong deviation in relation to other populations studied, suggesting the occurrence of a masking effect. The individual scores presented a positive correlation with the age (r = 0.31; p less than 0.01), and a significant intersexual difference was also verified (t = 3.08; p less than 0.01). This intersex difference is explained, in part, by analyzing the socioeconomic patterns of the community. The offspring/parent regression of the individual scores indicated a low dependency between genitors and their direct descendents (p greater than 0.7), and the estimative of heritability obtained (0.14) is artificial, since the offspring/mother and offspring/midparent regression coefficients were negative. Statistically non-significant coefficients of correlation and/or regression showed a highly randomic populational distribution of scores for the Horne-Ostberg's test. Such findings suggest that the intensity of the masking over the sleep/wake cycle varies among human populations, and that the individual tendency towards morningness/eveningness is strongly related to sociocultural factors.

Adolescent↗
PubMed2001

Do logistic regression and signal detection identify different subgroups at risk? Implications for the design of tailored interventions.

Identifying subgroups of high-risk individuals can lead to the development of tailored interventions for those subgroups. This study compared two multivariate statistical methods (logistic regression and signal detection) and evaluated their ability to identify subgroups at risk. The methods identified similar risk predictors and had similar predictive accuracy in exploratory and validation samples. However, the 2 methods did not classify individuals into the same subgroups. Within subgroups, logistic regression identified individuals that were homogeneous in outcome but heterogeneous in risk predictors. In contrast, signal detection identified individuals that were homogeneous in both outcome and risk predictors. Because of the ability to identify homogeneous subgroups, signal detection may be more useful than logistic regression for designing distinct tailored interventions for subgroups of high-risk individuals.

Adult↗
PubMed1999

Statistical model building and model criticism for human circadian data.

Mathematical models have played an important role in the analysis of circadian systems. The models include simulation of differential equation systems to assess the dynamic properties of a circadian system and the use of statistical models, primarily harmonic regression methods, to assess the static properties of the system. The dynamical behaviors characterized by the simulation studies are the response of the circadian pacemaker to light, its rate of decay to its limit cycle, and its response to the rest-activity cycle. The static properties are phase, amplitude, and period of the intrinsic oscillator. Formal statistical methods are not routinely employed in simulation studies, and therefore the uncertainty in inferences based on the differential equation models and their sensitivity to model specification and parameter estimation error cannot be evaluated. The harmonic regression models allow formal statistical analysis of static but not dynamical features of the circadian pacemaker. The authors present a paradigm for analyzing circadian data based on the Box iterative scheme for statistical model building. The paradigm unifies the differential equation-based simulations (direct problem) and the model fitting approach using harmonic regression techniques (inverse problem) under a single schema. The framework is illustrated with the analysis of a core-temperature data series collected under a forced desynchrony protocol. The Box iterative paradigm provides a framework for systematically constructing and analyzing models of circadian data.

Adult↗
PubMed1997

Comparison of 2-D calculations from periapical and occlusal radiographs versus 3-D calculations from CAT scans in determining bone support for cleft-adjacent teeth following early alveolar bone grafts.

OBJECTIVE: This investigation was conducted to determine the agreement between three-dimensional (3-D) calculations from CAT scans and two-dimensional (2-D) calculations from standard dental radiographs in evaluating bone support for cleft-adjacent teeth after primary bone grafting. DESIGN: This retrospective study utilized CAT scans and dental radiographs taken of the alveolar cleft in patients an average of 11 years after primary bone grafting. SETTING: The subjects were patients treated by the Cleft Palate Team at Children's Memorial Hospital and Loyola University Medical Center, Chicago, Illinois. PATIENTS: Fourteen UCLP patients (9 males, 5 females) agreed to participate in this study by undergoing CAT scan assessment of their alveolar cleft sites. They also had to have periapical or occlusal radiographs of the grafted cleft site taken within 6 months of the CAT scan. INTERVENTIONS: All patients underwent primary lip repair, placement of a passive palatal plate, primary alveolar bone grafting (mean age 6.4 months), and palatoplasty before 1 year of age. Major tooth movement through final orthodontics was completed by the time of the radiographic assessment. MAIN OUTCOME MEASURES: CAT scan sections were reformatted and reconstructed to three-dimensionally calculate the percentage of root covered by bone support for the 15 teeth adjacent to the graft cleft sites. Dental radiographs of the same teeth were also traced and digitized. Percentages of root supported by bone were also established using the dental radiographs by dividing the amount of root covered by bone, by the anatomic root length. RESULTS: A paired, two-sample t test revealed no significant differences between the two methods of assessment, while linear regression showed a statistically significant correlation between the CAT scan assessment and the percentages found on the radiographs. CONCLUSIONS: Routine dental radiographs were able to estimate the total 3-D bone support for the roots of cleft adjacent teeth as determined by CAT scan to a statistically significant degree when groups where compared. The clinical significance for evaluation of individual cases was less impressive with a wide range of variability and a level of agreement that required acceptance of differences up to 25%.

Adolescent↗
PubMed2005

Comparative study of three teicoplanin-based chiral stationary phases using the linear free energy relationship model.

Teicoplanin (T) is a macrocyclic glycopeptide that is highly effective as a chiral selector for enantiomeric separations. In this study, we used three teicoplanin-based chiral stationary phases (CSPs) - native teicoplanin, teicoplanin aglycon (TAG) and recently synthesized methylated teicoplanin aglycon (MTAG). In order to examine the importance of various interaction types in the chiral recognition mechanism the three related CSPs were evaluated and compared using a linear free energy relationship (LFER). The capacity factors of 19 widely different solutes, with known solvation parameters, were determined on each of the columns under the same mobile phase conditions used for the chiral separations. The regression coefficients obtained revealed the magnitude of the contribution of individual interaction types to the retention on the compared columns under those specific experimental conditions. Statistically derived standardized regression coefficients were used to evaluate the contribution of individual molecular interactions within one stationary phase. It has been concluded that intermolecular interactions of the hydrophobic type significantly contribute to retention on all the CSPs studied here. Other retention increasing factors are n- and pi-electron interactions and dipole-dipole or dipole-induced dipole ones, while hydrogen donating or accepting interactions are more predominant with the mobile phase than with the stationary phases. However, these types of interactions are not equally significant for all the CSPs studied.

Chromatography, High Pressure Liquid↗
PubMed1999

Further crossvalidation of regression-based neuropsychological norms with an update for the Boston Naming Test.

Regression-based norms for the Trail Making Test, Boston Naming Test, and Wisconsin Card Sorting Test, which we published in 1991 and 1993, have been criticized by Fastenau (1998) as having overcorrected for demographic influences in a sample of 63 older adults. We present data from new, independent participant samples that are consistent with expectations from the regression-based norms. We propose that Fastenau's findings in this instance resulted from the nonrepresentative nature of his relatively small sample, rather than from statistical deficiencies of regression based norms. Our currently published norms on one of the four tests considered here, the Boston Naming Test, are based upon a participant sample that was small and had inadequate representation of young adults. We address this by providing updated norms based on a much larger and more representative sample (N = 531).

Adult↗