Increasing the precision of data analysis: planned comparisons versus omnibus tests.
Explore the source record for details and available documents.
Analysis of Variance↗
SEARCH · PubMed Health
Results for “data analysis”
Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
[Conformation of the dienone system of delta 4,9-19-nor-3-ketosteroids from x-ray analysis data and its relation to reception and hormonal activity].
Basing on the data of the X-ray analysis of 2 beta-methylestr-4,9-dien-3-one-17 beta-ol (C19H26O2) and A-ring unsubstituted steroid 4,9-dien-3-ones, the noncomplanarity and flexibility of the conjugated dienone system in these transformed steroids has been demonstrated. The results have been also confirmed by UV spectroscopy. The ability of the dienone system to assume conformations with the out of- plane C3 = O3 and/or C9 = C10 bonds allows the AB-fragment of the steroid molecule to adopt the conformations required for interacting with various receptors. This property may also account for a simultaneous enhancement of several hormonal activities upon such a modification of steroids. The results of the X-ray analysis of 2 beta-methylestra-4,9-dien-3-one-17 beta-ol (space group P2(1)2(1)2(1), a 8,543(2), b 9,783(2), c 18,690(7) A, R 8,7%) are presented.
Ketosteroids↗
Statistics: 2. Data analysis.
Explore the source record for details and available documents.
Analysis of Variance↗
Good relations: the use of a relational database for large-scale data analysis.
This paper reports our experience of analysing what may well be one of the largest datasets gathered on nursing practice in the United Kingdom. The study produced both quantitative and qualitative data and a method had to be devised both for analysing each form of data and for relating the two. An inexpensive relational database was chosen for the purpose, and experience of using it is reported. Detailed examples are given. We look at the strengths and weaknesses of such a tool, and in general it received a positive evaluation. For many nursing research projects, it offers some advantages over a conventional statistical package, especially in the following areas: offering ease of use, and hence control of the data, by the domain (nursing) specialist; facilitating the analysis of free-text data; allowing the linking of free-text and structured questionnaire data; permitting the testing of conjectures which arise during analysis; handling varying amounts of data per case; providing non-redundant storage of data; permitting the association of machine-readable codes and human-readable labels; and encouraging an exploratory rather than merely analytical approach.
Data Interpretation, Statistical↗
Secondary data analysis: new perspective for adolescent research.
Adolescents are at high risk for preventable health problems, but realities of research funding and obtaining data from large samples make it difficult for many researchers to answer important questions. Secondary analysis of existing data sets is described as a reasonable alternative in spite of its limitations. Methods and available resources are identified along with suggestions for research priorities with adolescents.
Adolescent↗
Validation of large data sets, an essential prerequisite for data analysis: an analytical survey of the Bone Marrow Donors Worldwide.
Large data sets like the Bone Marrow Donors Worldwide (BMDW) data set can be used for population genetic analyses. The qualities of such data sets are unique. To be able to use the BMDW data for analyses, several problems, like limited size and selective DR typing, of the data have to be solved and the quality of the registry data subsets has to be examined. We describe these problems and methods to overcome them. Also, we give an overview of the qualities of the different registry subsets. Sixteen of the twenty-nine examined subsets contain data that can be used for population genetic analysis. We will deal with these analyses in the future. Additionally, we present a method to calculate the minimum number of individuals required for reliable haplotype frequency estimation.
Bone Marrow Transplantation↗
Exponential data-analysis of passive-avoidance behavior in rats and mice.
Data obtained with the passive-avoidance task are usually presented as the median values of the latencies to respond. In an earlier publication we described a better way of presenting such data based on the observation that the complement of the cumulative distribution of step-through latencies can be closely fitted by a simple exponential function. Thus the "step-through rate constant" (STRC) is concise and accurate quantitative description of population behavior in this test. In this paper we present two examples of the application of this procedure. In the first, variation in the interval between training and testing in rats changes the STRCs of the different groups. In the second (based on data published by Flood et al.) administration of cycloheximide is seen to partition the experimental population of mice into two subgroups with different STRCs.
Animals↗
A versatile physiological data analysis system using an Intel 8080 microprocessor.
A microprocessor based physiological data processor has been realised. The system controls the data flow from physiological experiments and performs on-line mean and variance calculations with an output in graphical form. The analyser accepts one data point every 0.5 ms and has a capacity of 128 records each containing 800 data points. Post stimulus histogram and interval histogram analysis programs have also been written and implemented.
Computers↗
Microarrays, pattern recognition and exploratory data analysis.
We explore the application of principal component based approaches to pattern recognition in microarray analysis. A comparative assessment is presented, based on predictive evaluation, following a review of some key methodology. On the basis of these results, we select principal component based linear discrimination for a further in-depth analysis. We are particularly interested in studying the use of principal component decomposition for the evaluation of the condition of estimation of the pooled covariance matrix. The results are used to give some guidance as to how principal components may be used as a data exploratory tool in the analysis of microarray data. Opportunities for further development are outlined as well as implications for wider statistical modelling of microarray data.
Computational Biology↗
Using voxel-specific hemodynamic response function in EEG-fMRI data analysis.
Most existing analytical techniques for EEG-fMRI data need specific assumptions about the hemodynamic response function (HRF). These assumptions may not be appropriate when the HRF varies from subject to subject or from region to region. In this article, we introduce a deconvolution method for EEG-fMRI activation detection, which can be implemented with voxel-specific HRFs. A comparison of performance is made between three fixed HRFs and the deconvolution method under the framework of the general linear model. The main results are as follows: (1) the volume of detected regions from the deconvolved HRFs is larger. (2) In some subjects, the deconvolution technique can find areas of activation that have not been detected with the three fixed HRFs at our threshold of significance. (3) Deconvolution obtained higher adjusted coefficients of multiple determination compared to those obtained with the three fixed HRFs. The results suggest that the fixed HRF methods may not be the most appropriate for the analysis of epileptic activity with EEG-fMRI, and the deconvolution method may be a better choice.
Brain↗
Clinical audit of patients with cervical cancer in Slovenia--data analysis for the year 2003.
PURPOSE OF INVESTIGATION: The data gathered in 2003 on the patients with cervical cancer who regularly attended their gynecologist were analyzed with the purpose of clinical audit. METHODS: The data on newly detected patients with cervical cancer in 2003 who regularly attended their gynecologist were gathered simultaneously at three Advisory Boards for Gynecology in Slovenia. RESULTS: Of 149 patients in whom, according to our data, invasive cervical cancer had been diagnosed, 92 (61.7%) patients were examined by a gynecologist in the previous five years. In the majority of these patients, cervical cancer was diagnosed in early, localized disease stage. In the periods of 13-24 and of seven to 12 months before the diagnosis of cervical cancer, almost half the patients had Pap II, and three to six months before diagnosis, 67.6% of patients had Pap II. CONCLUSION: These results encourage us to proceed with clinical audits to analyze individual cervical cancer cases, including another independent reevaluation of cervical smears in the five-year period before diagnosis. A suitable calendar of refresher training courses on colposcopy, which should be obligatory for all performing this examination method, also needs to be set up.
Adult↗
[The incidence and prevalence of malignant diseases. A secondary data analysis of a representative random sample of those with local compulsory medical insurance in Dortmund].
Prevalence and distribution of malignant neoplasms were obtained in 1988 and 1989 from a representative random sample of 5% of insured members of the statutory health insurance company in Dortmund, Germany (7447 persons). Data were gathered from medical certificates the diagnosis being further validated internally by other patient data. The cancer prevalence in the sample was 2.3% (males 1.6%, females 3.0%). The average age of the 174 patients with cancer was 67.5 years, half of them 70 years or over. The most frequent cancer was that of the breast (32.2%), followed by that of the colorectum (14.3%) and the skin (12.5%). Standardized for age and sex the prevalence of cancer for the whole population of the Federal Republic of Germany (excluding the former German Democratic Republic) was calculated as 2.2% (males 1.7%, females 2.5%), i.e. 1.33 million patients. The incidence of cancer for 1989 was 0.54%. Taking into account special methodological considerations, the secondary analysis of data obtained from the described source provides information relevant to the care of patients with cancer and thus makes a valuable contribution to the documentation of medical care provision.
Adult↗
[Geriatrics in Bavaria-Database (GiB-DAT): Basic data analysis of geriatric rehabilitation in Bavaria and analysis of inter-clinic variability (part II)].
OBJECTIVE: Presentation and comparison of basic data from geriatric rehabilitation in Bavaria with analysis of statistical influences on characteristic values like length of stay and Barthel Index. Are differences between participating clinics due to patient characteristics? SETTING: There were a total of 21 656 data records of the Geriatrics in Bavaria- Database (GiB-DAT) collected by 35 clinics within 1 year. RESULTS (PERCENTAGE OR MEAN): 70.6% female patients, age 80.0 years, 95.9% enrollment from acute care hospitals, length of stay in previous hospitals 24.3 days, length of stay in geriatric rehabilitation 24.6 days. DIAGNOSIS: 41.7% orthopedics, 24.9% neurological, 33.4% others. Barthel Index on admission 44.1 points, on discharge 65.8 points (difference 21.8 points). We found a large variance of all items between participating clinics. Variance for Barthel Index and length of stay is partly cleared up by regression analysis (multinomial logistic) and is mostly due to patient characteristics. CONCLUSIONS: Basic data of Bavarian geriatric rehabilitation clinics is matching with nationwide reference statistics. Differences exist for the lower rate of stroke and direct enrollment of outpatients. A simple ranking of outcome parameters (e. g. Barthel Index) does not make sense due to multiple influencing factors.
Aged↗
A naturalistic comparison of amoxicillin/clavulanate extended release versus immediate release in the treatment of acute bacterial sinusitis in adults: A retrospective data analysis.
BACKGROUND: Choosing the most effective treatment for acute bacterial sinusitis (ABS) is helpful to avoid treatment failure. To date, studies comparing antibiotic options for ABS have compared amoxicillin/clavulanate (AMX/CA) immediate release (IR) versus other antibiotics, but have not included AMX/CA extended release (XR). OBJECTIVE: The purpose of this study was to determine whether the clinical advantages of AMX/CA XR found in a clinical trial setting translate to a naturalistic setting, relative to AMX/CA IR. METHODS: Data for this retrospective analysis were obtained from a managed care benchmark database that included >25 million patients from >30 health plans covering 7 US census divisions. Data from all patients aged > or =18 years with an index diagnosis of ABS between July 1, 2001, and December 31, 2003, were included. Episodes of ABS were classified as treatment successes (no additional prescriptions for antibiotics, ABS-related emergency department [ED] visits, or ABS-related inpatient hospitalizations within 30 days after the index prescription) or failures (> or =1 subsequent antibiotic prescription, an ABS-related ED visit, or an ABS-related inpatient hospitalization within 30 days after the index prescription). Treatment failures were subclassified as early or late. Mean costs were assessed for medical claims linked with a primary diagnosis of ABS and antibiotic pharmacy claims within the follow-up time period (through February 28, 2004). Descriptive statistics for demographic characteristics, utilization patterns, and success rates were calculated for each cohort. A multivariate general linear regression model was developed to assess differences in costs between the 2 cohorts. RESULTS: : Of the patients with an index antibiotic prescription filled within 3 days of ABS diagnosis (n = 241,511), a total of 3224 in the AMX/CA XR cohort (mean [SD] age, 41.8 [11.6] years; women, 57.7%) and 23,638 in the AMX/CA IR cohort (mean [SD] age, 41.9 [11.6] years; women, 62.7%) were included in the analysis. The rate of treatment success in patients treated with AMX/CA XR was 82.8% versus 81.0% in patients treated with AMX/CA IR (P < 0.015). Treatment success costs were significantly lower than treatment failure costs regardless of an early or late designation and ranged from 98 dollars to 110 dollars per episode (P < 0.001). After adjustment for background covariates, the mean cost of treating an episode of ABS was significantly lower for patients receiving AMX/CA XR versus those receiving AMX/CA IR (166.32 dollars vs 177.34 dollars [US 2004 dollars]; P < 0.001), representing a mean cost savings of 11.02 dollars per patient treated with AMX/CA XR over AMX/CA IR for ABS, regardless of treatment outcome. CONCLUSIONS: The results from this data analysis suggest that AMX/CA XR had significantly higher treatment success in ABS relative to AMX/CA IR in this naturalistic setting. AMX/CA XR was associated with significantly decreased total ABS-related costs in these adults.
Acute Disease↗
The effect of normalization on microarray data analysis.
This paper contains a description of several common normalization methods used in microarray analysis, and compares the effect of these methods on microarray data. The importance of background subtraction is also addressed. The research focuses on three parts. The first uses three statistical methods: t-test, Wilcoxon signed rank test, and sign test to measure the difference between background subtracted data and nonbackground subtracted data. The second part of the study uses the same three statistical methods to compare whether data normalized with different normalization methods yield similar results. The third part of the study focuses on whether these differently normalized data will influence the result of gene selection (dimension reduction). The comparisons are done for several data sets to help identify similarity patterns. The conclusion of this study is that background subtraction can make a difference, especially for some data sets with poorer quality data. The choice of normalization method, for the most part, makes little difference in the sense that the methods produce similarly normalized data. But, based on the third part of analysis, we found that when gene selection is performed on these differently normalized data, somewhat different gene sets are obtained. Thus, the choice of normalization method will likely have some effect on the final analysis.
Oligonucleotide Array Sequence Analysis↗
Association between congestive heart failure and hospitalization in patients with type 2 diabetes mellitus receiving treatment with insulin or pioglitazone: a retrospective data analysis.
BACKGROUND: Thiazolidinediones (TZDs) are widely used oral antihyperglycemic drugs that facilitate insulin action and increase insulin-stimulated glucose metabolism, thereby decreasing insulin resistance. However, concerns have been raised regarding the association between TZD use and a heightened risk for congestive heart failure (CHF). OBJECTIVE: This study used claims data to conduct a retrospective examination of the CHF risk in patients with type 2 diabetes mellitus and to compare the association with CHF in those receiving the TZD pioglitazone and those receiving insulin. METHODS: Patients with type 2 diabetes aged > or =18 years who had begun treatment with pioglitazone or insulin between January 1999 and December 2001 were identified using the PharMetrics Patient-Centric database. The sample was restricted to patients for whom there were > or =12 months of data before the index date (date of the first prescription for pioglitazone or insulin) and > or =3 months of follow-up data. Patients receiving a diagnosis of CHF before the index date were excluded. The propensity score for receiving pioglitazone was estimated using logistic regression based on available observed patient characteristics. Patients receiving insulin were matched in a 1:1 ratio with patients receiving pioglitazone based on a difference of no more than +/-0.01 in the estimated propensity score for receiving pioglitazone therapy. CHF risk was examined using the Cox proportional-hazards model. RESULTS: After exclusion of ineligible patients, 1668 matched pairs of patients receiving pioglitazone or insulin were identified (50.9% men, 49.1% women; mean [SE] age, 51.2 [0.2] years). The 2-year crude incidence rate of CHF was significantly lower in the pioglitazone group compared with the insulin group (primary/secondary diagnosis of CHF, 2.0% vs 4.0%, respectively; P < 0.001; inpatient hospitalization for CHF, 0.7% vs 2.5%; P < 0.001). The hazard ratio for pioglitazone versus insulin was 0.501 (95% CI, 0.331-0.758; P = 0.001) for a primary or secondary diagnosis of CHF in any setting and 0.263 (95% CI, 0.135-0.511; P < 0.001) for any occurrence of an inpatient hospitalization for CHF. CONCLUSIONS: In this retrospective analysis of data from patients with type 2 diabetes, pioglitazone therapy was associated with significantly lower incidence rates of CHF and inpatient hospitalization compared with insulin therapy.
Cohort Studies↗
Moving in a given year: a study in research design and data analysis.
"In this paper the distribution of the times between moves, or, equivalently, the relationship between duration of residence and the probability of moving [is examined]. Complete information on residence histories contained in the Population Register of Norway for the period 1965-1971 has been used to generate two data sets: the 'ideal' data set and the duration of residence at a specific point in time." The relationship between the probability of moving in a given year and stochastic models of mobility histories is then discussed. "In the subsequent section the feasibility of using the probability of moving in a given year, as measured by cross-sectional survey techniques, in lieu of complete residence histories, as contained in population registers, is evaluated. It could be demonstrated that cross-sectional data is sufficient for many purposes." In conclusion, "some comments are made regarding methodological problems arising out of the use of sparse tabular data, which is typical in socio-demographic studies where highly disaggregated projections are required."
Demography↗
Develop data analysis before signing risk contract to ensure maximum savings.
Seeking a risk contracting edge, a Pennsylvania hospital-physician alliance turns to data gathering and analysis to reduce costs, improve care, and get the most out of its contract. Leaders of PennCARE, an alliance of 11 hospital systems and multiple physician groups, are spending a lot of time and energy these days cleaning up and analyzing payer claims data. Find out what they look for to reduce variation in care among the alliance's hospitals and physicians. And see how providers are using data to improve office processes, as well.
Benchmarking↗