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 955 records · Page 53Linked to original sources
PubMed1995

High fasting insulin levels associated with lower rates of weight gain in persons with normal glucose tolerance: the San Luis Valley Diabetes Study.

OBJECTIVE: to investigate whether increased insulin resistance is associated with a reduced risk of weight gain among Hispanic and non-Hispanic white persons in Colorado. DESIGN: measurements were taken at baseline and after an average of 4.3 years. SUBJECTS: 789 normal glucose tolerant subjects 20 to 74 years of age were studied. MEASUREMENTS: fasting insulin levels were used as an indirect marker of insulin resistance. RESULTS: the average baseline body mass index (BMI) was 26.1 (+/- 3.6 s.d.) in men and 25.2 (+/- 4.4 s.d.) in women. The average weight change at follow-up was 0.8 kg (+/- 4.3 s.d.) in men and 1.3 kg (+/- 5.0) in women. A doubling in initial fasting insulin was associated with a reduced risk of gaining 5 kg (n = 134) and 10 kg (n = 24) of weight in logistic regression models (OR5kg = 0.76, 95% CI: 0.59, 0.99 and OR10kg = 0.61, 95% CI: 0.36, 1.02). Similarly, a doubling of initial fasting insulin was associated (P = 0.006) on average with a 6.3 kg less weight gain in linear regression models, independently of initial body weight, age, sex, ethnicity and BMI. The relation was consistently observed in men and women and in both ethnic groups. CONCLUSION: higher initial fasting insulin decreases the risk of subsequent weight gain in both Hispanic and non-Hispanic white normal glucose tolerant individuals similar to Pima Indians. This appears to be a common biologic characteristic in moderate to low as well as high risk populations for NIDDM.

Adult↗
PubMed1993

Ultrasonographic determination, in vitro and in vivo, of canine gallbladder volume, using four volumetric formulas and stepwise-regression models.

Twelve resected canine gallbladders (in vitro) and the gallbladder in each of 14 dogs (in vivo) were ultrasonographically examined. Gallbladder volume was calculated from ultrasonographically measured geometric dimensions, using 4 volumetric model formulas: cone, ellipse, biplanar ellipse, and prolate ellipse. Calculated volume was compared with true gallbladder volume, as measured by water displacement. All examined models for calculation of gallbladder volume were closely associated with true gallbladder volume (P < 0.005), and all models provided accurate predictions of true gallbladder volume (r2 > 0.80). Calculated volumes can be corrected mathematically by use of the regression coefficient and constant for each model. Body weight was not significantly associated with gallbladder volume in any of the models considered. Use of ultrasonography to accurately measure gallbladder volume could be combined with synthetic cholecystokinin-stimulated gallbladder emptying to provide information about biliary function and patency in icteric animals. Such information could aid the clinical decision between surgical or medical treatment. Correction of calculated volumes would not be necessary in association with induced emptying studies, because volume change is more important than absolute volume.

Animals↗
PubMed1994

An investigation into factors that affect oesophageal temperature during abdominal surgery.

The purpose of this study was to examine factors which influence changes in body temperature during surgical operation. Variables included were ambient theatre temperature and humidity, duration of surgery, age of patient, index of body fat, blood loss during surgery, and the volume of intravenous fluids received. Subjects were 30 female patients undergoing elective abdominal gynaecological surgery. Body temperature was measured using oesophageal probes and data were collected for the duration of the operation. Patients sustained a mean reduction in oesophageal temperature of 1.03 degree C. Bivariate correlations indicated that ambient temperature, age, body fat, volume of intravenous fluids and blood loss were associated with temperature reduction at the level of < 0.05. Step-wise multiple regression revealed two possible models including ambient theatre temperature, each of which accounted for 70% of the variance. Model 1 included age, while model 2 included volume of intravenous fluids and index of body fat. It was noted that older patients were generally in theatre at a time when the ambient temperature was lowest. Recommendations are made to minimize unnecessary reductions in body temperature during surgery for vulnerable groups, notably elderly and thin patients.

Adipose Tissue↗
PubMed1997

Estimating benchmark concentrations and other noncancer endpoints in epidemiology studies.

Methods for evaluating the hazards associated with noncancer responses with epidemiologic data are considered. The methods for noncancer risk assessment have largely been developed for experimental data, and are not always suitable for the more complex structure of epidemiologic data. In epidemiology, the measurement of the response and the exposure is often either continuous or dichotomous. For a continuous noncancer response modeled with multiple regression, a variety of endpoints may be examined: (1) the concentration associated with absolute or relative decrements in response; (2) a threshold concentration associated with no change in response; and (3) the concentration associated with a particular added risk of impairment. For a dichotomous noncancer response modeled with logistic regression, concentrations associated with specified added/extra risk or with a threshold responses may be estimated. No-observed-effect concentrations may also be estimated for categorizations of exposures for both continuous and dichotomous responses but these may depend on the arbitrary categories chosen. Respiratory function in miners exposed to coal dust is used to illustrate these methods.

Age Factors↗
PubMed1991

Psychosocial and behavioral factors associated with risk of sexually transmitted diseases, including human immunodeficiency virus infection, among urban high school students.

The main purpose of this study was to evaluate the role of multiple psychosocial and knowledge-related antecedent factors that may predict sexual and alcohol and drug use behaviors that are associated with the transmission of sexually transmitted diseases (STDs), including human immunodeficiency virus infection. Five hundred forty-four ninth-grade urban high school students were surveyed regarding knowledge, attitudes, and behaviors related to STDs and acquired immunodeficiency syndrome (AIDS). Results of multiple regression analyses indicate that higher levels of STD and AIDS knowledge were associated with lower levels of STD and AIDS anxiety (R2 delta = 0.09; p less than 0.001), fewer negative attitudes toward people with AIDS (R2 delta = 0.09; p less than 0.001), stronger perceptions of self-efficacy (R2 delta = 0.03; p less than 0.01), and stronger peer affiliation (R2 delta = 0.02; p less than 0.05). Negative attitudes toward people with AIDS were inversely related to knowledge (R2 delta = 0.08; p less than 0.001), social support (R2 delta = 0.02; p less than 0.01), and perceived self-efficacy (R2 delta = 0.01; p less than 0.05). Predictors of alcohol and drug use included perceived peer norms (R2 delta = 0.08; p less than 0.001) and strong peer affiliation (R2 delta = 0.05; p less than 0.001). The best predictor of sexual risk behavior was alcohol and drug use (R2 delta = 0.07; p less than 0.001). Lower levels of knowledge (R2 delta = 0.14; p less than 0.01) and perceived peer norms (R2 delta = 0.05; p less than 0.05) predicted nonuse of condoms. Our results indicate that several factors relate to adolescent risk for STDs: the connection between peer influence and adolescent risk behaviors, the link between alcohol and drug use and sexual risk behavior, and the role of knowledge in determining nonuse of condoms.

Acquired Immunodeficiency Syndrome↗
PubMed2003

SNP subset selection for genetic association studies.

Association studies for disease susceptibility genes rely on the high density of SNPs within candidate genes. However, the linkage disequilibrium between SNPs imply that not all SNPs identified in the candidate region need be genotyped. Here we develop several approaches to SNP subset selection, which can substantially reduce the number of SNPs to be genotyped in an association study. We apply clustering algorithms to pairwise linkage disequilibrium measures, with SNP subsets determined for different cut-off values of Delta using nearest and furthest neighbour clusters. Alternatively, SNP subsets may be determined by the proportion of haplotypes they identify. We also show how power calculations, based on the average power to identify a SNP as the disease susceptibility mutation using haplotype-based or logistic regression based statistical analyses, can be used to choose SNP subsets. All these methods provide a ranking method for subsets of a specific size, but do not provide criteria for overall choice of SNP subset size. We develop such criteria by incorporating power calculations into a decision analysis, where the choice of SNP subset size depends on the genotyping costs and the perceived benefits of identifying association. These methods are illustrated using eleven SNPs in the MMP2 gene.

Cluster Analysis↗
PubMed2007

Cervical spine curvature and craniofacial morphology in an adult Caucasian group: a multiple regression analysis.

The aim of this study was to investigate the relationship between cervical curvature and skeletal facial morphology. Pre-treatment lateral cephalograms in natural head position were obtained from 98 orthodontic adult subjects (56 males and 42 females; mean age = 31.5 years, standard deviation +/-5.8 years). The amount of concavity of the cervical curvature was calculated by a second-order quadratic interpolation of the most infero-posterior points of the bodies of the seven cervical vertebrae on the radiographs. Linear regression with stepwise elimination was carried out to evaluate the sagittal and vertical craniofacial dimensions. The results revealed that the amount of curvature was related to (1) the horizontal position of the upper incisors with respect to the maxillary and mandibular skeletal bases and (2) the sagittal millimetric dimension of the maxillary base. These cephalometric variables explained 51 per cent of the total variance of cervical curvature. There was no significant correlation between cervical curvature and any other cephalometric variable. Cervical curvature was not influenced by age or gender.

Adolescent↗
PubMed1995

Validation of body condition scores with ultrasound measurements of subcutaneous fat of dairy cows.

This research validated body condition scores with ultrasound measurements of subcutaneous fat. Fifty Holstein cows were evaluated during three sessions in 1993. Cows scored during each session were divided into three groups of 15 or 20 cows. Body condition scores were assigned by one trained individual, utilizing a five-point (1 = thin to 5 = fat) visual scoring technique. Cows were scored to the nearest quarter point. Ultrasound measurements of subcutaneous fat were obtained by another individual at the lumbar, thurl, and tailhead areas of both sides of the cow. Body condition scores and ultrasound measurements were collected on the same day, but obtained independently. Correlations between ultrasound measurements ranged from .36 to .86. Regression models were developed to validate the body condition scoring technique across the three cow groups. Group number and different combinations of ultrasound measurements were independent variables, and body condition score was the dependent variable. Ultrasound measurements were significantly associated with body condition scores. The coefficients of determination for the models ranged from .36 to .65, depending on which ultrasound measurements were included in the model. These results suggest that the body condition scoring technique used in this study was as valid as ultrasound techniques for measurement of subcutaneous fat.

Adipose Tissue↗
PubMed1992

Contraceptive compliance with a levonorgestrel triphasic and a norethindrone monophasic oral contraceptive in adolescent patients.

OBJECTIVE: This study was undertaken to assess the impact of two low-dose oral contraceptive pills on compliance and side effects in adolescent patients. STUDY DESIGN: The use of a levonorgestrel-containing triphasic pill (N = 114) was compared with that of a monophasic (1 + 35) norethindrone-containing pill (N = 110) at two different sociodemographic sites. RESULTS: No significant difference in compliance or pill satisfaction was observed between the pills. Socioeconomic factors were the overriding predictors of compliance. At 3 and 12 months of follow-up, there were significantly fewer complaints of overall side effects (p less than 0.001 and p = 0.004, respectively), breakthrough bleeding (p = 0.017 and p = 0.018), and pill amenorrhea (p = 0.002 and p less than 0.001) among users of the triphasic pill. Mean weight change at 12 months was +1.1 kg for the monophasic pill and -0.1 kg for the triphasic pill. All known pregnancies occurred among noncompliant city clinic patients. CONCLUSIONS: Adolescents experienced fewer side effects with the triphasic pill than with the monophasic one, but compliance was the same.

Adolescent↗
PubMed2003

2-[11C]thymidine positron emission tomography as an indicator of thymidylate synthase inhibition in patients treated with AG337.

BACKGROUND: Some anticancer drugs inhibit thymidylate synthase (TS), a key enzyme for thymidine nucleotide biosynthesis. Cells can compensate for depleted thymidine levels by taking up extracellular thymidine via a salvage pathway. We investigated the use of 2-[11C]thymidine positron emission tomography (PET) to measure thymidine salvage kinetics in vivo in humans. METHODS: Five patients with advanced gastrointestinal cancer were PET scanned both before and 1 hour after oral administration of the TS inhibitor AG337 (THYMITAQ [nolatrexed]); seven control patients were scanned twice but not treated with AG337. Thymidine salvage kinetics were measured in vivo using 2-[11C]thymidine PET and spectral analysis to obtain the standardized uptake values (SUV), the area under the time-activity curve (AUC), and the fractional retention of thymidine (FRT). Changes in PET parameters between scans in the AG337-treated and control groups were compared using the Mann-Whitney U test. The relationship between AG337 exposure and AG337-induced changes in tumor FRT and in plasma deoxyuridine levels (a conventional pharmacodynamic systemic measure of TS inhibition) was examined using Spearman's regression analysis. Statistical tests were two-sided. RESULTS: The between-scan change in FRT in patients treated with AG337 (38% increase, 95% confidence interval [CI] = 8% to 68%) was higher than that in control patients (3% increase, 95% CI = -11% to 17%) (P =.028). The level of AG337-induced increase in both 2-[11C]thymidine FRT and plasma deoxyuridine levels was statistically significantly correlated with AG337 exposure (r = 1.00, P =.01 for both). CONCLUSIONS: AG337 administration was associated with increased tumor tracer retention that was consistent with tumor cell uptake of exogenous 2-[11C]thymidine as a result of TS inhibition. 2-[11C]Thymidine PET can be used to measure thymidine salvage kinetics directly in the tissue of interest.

Adult↗
PubMed2003

The Finno-Ugrian suicide hypothesis: variation in European suicide rates by latitude and longitude.

The marked variation regarding the suicide rate in 34 European countries is well described by regressing the national suicide rate on the capital cities' latitudes and on an interaction term of squared latitude multiplied with longitude. The interaction term explains 40.8% and 29.1% of men's and women's suicide rate, respectively, and latitude explains a further significant increment of 10.9% and 10.6% variance of men's and women's suicide rate, respectively. This regression model quantifies the Finno-Ugrian suicide hypothesis of Kondrichin and of Marusic and Farmer. The European countries highest in suicide rate constitute a contiguous, J-shaped belt, spanning from Finland to Austria. This area maps onto the second principal component identified for European gene distribution, representing ancestral adaptation to cold climates and the Uralic language dispersion. Thus, population differences in genetic risk factors may account for the spatial pattern in European suicide rates.

Acclimatization↗
PubMed1993

Estimating shelf-life using L1 regression methods.

The shelf-life of a drug is usually determined by fitting the assayed values of drug potency versus time to a linear least-squares model. In least-squares models, the parameters are estimated by minimizing the sum of squares of the residuals. Shelf-life is estimated as the time corresponding to the intersection point of the fitted line and the minimum acceptable per cent or proportion of the initial drug potency. The least-squares method can be seriously affected by outliers, leading to erroneous shelf-life estimates. In this paper, an alternative method, based on minimizing the sum of absolute deviations, was applied to the shelf-life determination problem. The resistance of these L1 based estimates to outliers was demonstrated using a typical stability dataset.

Drug Stability↗
PubMed2005

An investigation of home advantage and other factors affecting outcomes in English one-day cricket matches.

We examined the factors affecting the outcome of cricket matches played in the English one-day county cricket league. In particular, we focused on the home-field effect and the importance of winning the pre-match toss of a coin to determine a team's strategic decision to bat first or second. A home-field effect appeared to be confirmed in that home teams won 57% of all matches with a win/loss result. A logistical regression model was used, with the outcome variable defined in terms of a home team win/loss. We found that while winning the toss is an important aspect of a one-day cricket match, other factors tend to dominate in determining the result, especially team quality and match importance for the home and away teams in the overall league context. Our results also indicate, not surprisingly given the nature of cricket attendance and spectating, that the crowd effect is largely insignificant. The results of our study do not support any rule changes requiring the abandonment of the coin toss to determine batting order.

Competitive Behavior↗
PubMed2001

Akaike's information criterion in generalized estimating equations.

Correlated response data are common in biomedical studies. Regression analysis based on the generalized estimating equations (GEE) is an increasingly important method for such data. However, there seem to be few model-selection criteria available in GEE. The well-known Akaike Information Criterion (AIC) cannot be directly applied since AIC is based on maximum likelihood estimation while GEE is nonlikelihood based. We propose a modification to AIC, where the likelihood is replaced by the quasi-likelihood and a proper adjustment is made for the penalty term. Its performance is investigated through simulation studies. For illustration, the method is applied to a real data set.

Biometry↗
PubMed2004

Asthma disease management: regression to the mean or better?

OBJECTIVES: To assess the effectiveness of disease management as an adjunct to treatment for chronic illnesses, such as asthma, and to evaluate whether the statistical phenomenon of regression to the mean is responsible for many of the benefits commonly attributed to disease management. STUDY DESIGN: This study evaluated an asthma disease management intervention in a Colorado population covered by Medicaid. The outcomes are presented with the intervention group serving as its own control (baseline and postintervention measurements) and are compared with a matched control group during the same periods. METHODS: In the intervention group, 388 asthmatics entered and 258 completed the 6-month program; 446 subjects participated in the control group. Facilities charges were compared for both groups during the baseline and program periods. Both groups were well matched demographically and for costs at baseline. RESULTS: Using the intervention group as its own control revealed a 49.1% savings. The control group savings were 30.7%. Therefore, the net savings were 18.4% (P < .001) for the intervention group vs controls. Although the demonstrated savings were less using a control group to correct for regression to the mean, they were statistically significant and clinically relevant. CONCLUSION: When using a control group to control for the statistical effects of regression to the mean, a disease management intervention for asthma in a population covered by Medicaid is effective in reducing healthcare costs.

Adolescent↗
PubMed1990

The relation of chlamydial infection of the cervix to time elapsed from the onset of menses.

To investigate the relation of chlamydial infection of the cervix to time elapsed from the onset of menses, we prospectively studied 338 women receiving routine gynecologic care in two clinical sites in which the prevalence of chlamydial infection was 3 and 6%. Eleven (13%) of 86 women evaluated in the fourth or fifth week after the onset of menses (days 22-35) had chlamydial infection compared to 4 (1.6%) of 252 women seen in the first 3 weeks (days 1-21) (p less than 0.001). For women seen in the fourth or fifth week, the relative risk of chlamydial infection was 8.1 (95% CI, 3.1-20.6). Controlling for known risk factors for chlamydial infection with multiple logistic regression analysis, the relative risk was sustained (RR 10.8; 95% CI, 3.1-37.4). We conclude that time elapsed from the onset of menses is an important independent risk factor for chlamydial infection of the cervix that may lead to more effective and efficient screening strategies.

Adult↗
PubMed2003

The dominance analysis approach for comparing predictors in multiple regression.

A general method is presented for comparing the relative importance of predictors in multiple regression. Dominance analysis (D. V. Budescu, 1993), a procedure that is based on an examination of the R2 values for all possible subset models, is refined and extended by introducing several quantitative measures of dominance that differ in the strictness of the dominance definition. These are shown to be intuitive, meaningful, and informative measures that can address a variety of research questions pertaining to predictor importance. The bootstrap is used to assess the stability of dominance results across repeated sampling, and it is shown that these methods provide the researcher with more insights into the pattern of importance in a set of predictors than were previously available.

Algorithms↗
PubMed1990

Ecological association between hypertension and stroke in Catalonia (Spain): development and use of an ecological regression model.

The objectives of this paper were to study the association between the prevalence of uncontrolled hypertension (PUHT) and stroke mortality at the ecological level, in nine geographical areas of Catalonia (Spain); to develop an ecological regression model and to assess its ability to predict crude stroke mortality rate (CMR) from the PUHT. The regression equation obtained for the population older than 25 yrs was CSMR x 10(3) = 0.67035 + 4.94752 PUTH x 10(-2). The ecological risk ratio was 8.38 and the ecological attributable proportion 71.1%. The CSMR estimation obtained by applying the model in a concrete case was close to that observed. The results support an ecological association between the studied variables. The ecological model can be useful in the assessment of observed changes in health problems and risk factor levels in the community. It could also be used in the evaluation of intervention programmes.

Cerebrovascular Disorders↗