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 1,009 records · Page 56Linked to original sources
PubMed2004

Flexible meta-regression functions for modeling aggregate dose-response data, with an application to alcohol and mortality.

In this paper, the authors describe fractional polynomials and cubic splines with which to represent smooth dose-response relations in summarizing meta-analytical aggregate data. Use of these two curve-fitting families can help prevent the problems arising from inappropriate linearity assumptions. These methods are illustrated in the problem of estimating the shape of the dose-response curve between alcohol consumption and all-cause mortality risk. The authors considered aggregate data from 29 cohort studies investigating this issue (1966-2000). J-shaped curves with a nadir at approximately 5-7 g/day of alcohol consumption and a last protective dose of 47-60 g/day were consistently obtained from fractional polynomials and cubic splines. The authors conclude that both of the curve-fitting families are useful tools with which to explore dose-response epidemiologic questions by means of meta-analytical approaches, especially when important nonlinearity is anticipated.

Alcohol Drinking↗
PubMed1992

Changing patterns in reported sexual practices in the population: multiple partners and condom use.

OBJECTIVE: To determine whether the pattern of change with regard to condom use and multiple sexual partners is influenced by gender or educational level. DESIGN: Findings from data collected from 1987 to 1990 about changes in condom use and multiple-partner activities are presented, based on telephone interviews with 9416 participants aged 18 to 44 years resident in central Scotland, UK. METHODS: Change in patterns over time were modelled in a multivariate logistic regression using a linear interactive modelling program. RESULTS: Several models showing changes in the proportion of multiple-partner respondents and condom users yielded a complicated pattern of behavioural change in educational status and gender. CONCLUSIONS: There is a large difference in reported condom use and multiple sexual partners by gender, but the difference is decreasing over time. Better educated respondents increased their use of condoms while less educated respondents showed a decrease in the proportion of multiple partners.

Adolescent↗
PubMed2006

Reducing substance use improves adolescents' school attendance.

AIMS: Substance use initiation and frequency are associated with reduced educational attainments among adolescents. We examined if decreases in substance use substantially improve youths' school attendance. DESIGN: A total of 1084 US adolescents followed quarterly for 1 year after entering substance abuse treatment. METHODS: Random and fixed effects regression models were used to differentiate the lagged effects of drug use from other time-varying and time-invariant covariates. Self-reports of alcohol, marijuana, stimulants, sedatives, hallucinogens and other drug use were used to predict subsequent school attendance, after controlling for demographic and drug use history characteristics, problem indices and other covariates. FINDINGS: Reductions in the frequency of alcohol, stimulants and other drug use and the elimination of marijuana use were each associated independently with increased likelihoods of school attendance. CONCLUSIONS: Because years of completed schooling is highly correlated with long-term social and economic outcomes, the possibility that reductions in substance use may improve school attendance has significant implications for the cost-effectiveness of substance abuse treatment and other interventions designed to reduce adolescents' substance use.

Absenteeism↗
PubMed2002

Comparing contrast-enhanced ultrasound to immunohistochemical markers of angiogenesis in a human melanoma xenograft model: preliminary results.

This study compared contrast-enhanced ultrasound (US) measures of tumor neovascularity with molecular markers of angiogenesis in a human melanoma xenograft model. A total of 14 mice were implanted with a human melanoma cell line (WM-9) in the thigh. After 2 to 3 weeks, a tumor, approximately 12 mm in diameter, developed. The US contrast agent Optison (Mallinckrodt, St. Louis, MO) was injected in a tail vein (dose: 0.4 to 0.6 mL/kg). Power Doppler and pulse-inversion harmonic imaging (HI) were performed with an Elegra scanner (Siemens Medical Systems, Issaquah, WA) and a 7.5 MHz linear array. Frame-rates of 30 Hz and 0.5 Hz (intermittent imaging) were used for pulse-inversion HI. After surgical removal, specimens were sectioned in the same planes as the US images. Immunohistochemical stains for endothelial cells (CD31), vascular endothelial growth factor (VEGF), basic fibroblast growth factor (bFGF) and cyclooxygenase-2 (COX-2) were performed. Two observers graded the stains (for intensity and percent area), and two other observers graded the US imaging modes (for fractional tumor neovascularity) on the same scale from 0 to 3. Of the 14 mice, 4 failed for technical reasons (i.e., n = 10). Linear regressions indicated statistically significant correlations between percent area stained with COX-2 and power Doppler (r = -0.789; p < 0.01), as well as intermittent pulse-inversion HI (r = -0.795; p < 0.05). There was a trend toward significance between percent area stained with VEGF and intermittent pulse-inversion HI (r = -0.720; 0.05 < p < 0.10). No other comparisons were significant. In conclusion, contrast-enhanced US measures of tumor neovascularity in a human melanoma xenograft model appear to provide a noninvasive marker of angiogenesis corresponding to expression of COX-2. However, the sample size of this study is small and, until further studies have been conducted, these conclusions are preliminary.

Albumins↗
PubMedDate not supplied

Fertility determinants in the oil region of Mexico.

This study analyzes fertility determinants in the oil region of Mexico, consisting of the states of Veracruz, Tabasco, and Campeche. Data are from the 1980 Mexican census and the unit of analysis is the municipio. The regression models, in which the dependent variables of children ever born and child-woman ratio are examined, reveal religious variables to be most significant, with greater fertility for non-Catholics and persons with no religion than for Catholics. Also of great importance are economic variables. Literacy and urbanization, both "classical" Mexican fertility variables, reduce fertility. There are major differences among three urban/rural and three indigenous language subsamples. Results are discussed vis-a-vis demographic theories and prior research.

Adolescent↗
PubMed1978

Correlates of contraceptive behavior among unmarried U.S. college students.

The contraceptive behavior of 213 sexually active never-married U.S. college students was examined using multiple regression analyses. The type of relationship--casual or steady--was a major determinant of contraceptive use: 77 percent of the respondents involved in steady relationships reported using condoms or the pill, compared with 58 percent of the respondents involved in casual relationships. Frequency of intercourse was the best predictor of contraceptive method used within steady relationships. Several variables emerged as predictors of contraceptive method used within casual relationships, including method used at first coital experience, age, and frequency of intercourse. Situational and background factors were found to have an important impact on adolescent contraceptive behavior and to warrant further research.

Adolescent↗
PubMed2005

A brief conceptual tutorial on multilevel analysis in social epidemiology: investigating contextual phenomena in different groups of people.

STUDY OBJECTIVE: (1) To provide a didactic and conceptual (rather than mathematical) link between multilevel regression analysis (MLRA) and social epidemiological concepts. (2) To develop an epidemiological vision of MLRA focused on measures of health variation and clustering of individual health status within areas, which is useful to operationalise the notion of "contextual phenomenon". The paper shows how to investigate (1) whether there is clustering within neighbourhoods, (2) to which extent neighbourhood level differences are explained by the individual composition of the neighbourhoods, (3) whether the contextual phenomenon differs in magnitude for different groups of people, and whether neighbourhood context modifies individual level associations, and (4) whether variations in health status are dependent on individual level characteristics. DESIGN AND PARTICIPANTS: Simulated data are used on systolic blood pressure (SBP), age, body mass index (BMI), and antihypertensive medication (AHM) ascribed to 25 000 subjects in 39 neighbourhoods of an imaginary city. Rather than assessing neighbourhood variables, the paper concentrated on SBP variance between individuals and neighbourhoods as a function of individual BMI. RESULTS: The variance partition coefficient (VPC) showed that clustering of SBP within neighbourhoods was greater for people with a higher BMI. The composition of the neighbourhoods with respect to age, AHM use, and BMI explained about one fourth of the neighbourhood differences in SBP. Neighbourhood context modified the individual level association between BMI and SBP. Individual level differences in SBP within neighbourhoods were larger for people with a higher BMI. CONCLUSIONS: Statistical measures of multilevel variations can effectively quantify contextual effects in different groups of people, which is a relevant issue for understanding health inequalities.

Adult↗
PubMed1990

Pregnancy, breast feeding, and oral contraceptives and the risk of epithelial ovarian cancer.

To quantify the effects of cumulative months of pregnancy, breast feeding, and oral contraceptive use on the risk of developing epithelial ovarian cancer, the authors used data collected for the Cancer and Steroid Hormone Study--a multicenter, population-based, case-control study. Detailed reproductive histories were obtained from 436 women aged 20-54 with epithelial ovarian cancer newly diagnosed between December 1980 and December 1982, and from 3833 women aged 20-54 selected at random from the same geographic areas. Estimated relative risks of epithelial ovarian cancer were 0.6 (95% confidence interval (CI) 0.5-0.8) for women who had ever been pregnant, 0.6 (95% CI 0.5-0.8) for women who had ever breast fed, and 0.5 (95% CI 0.5-0.7) for women who had ever used oral contraceptives. Logistic regression analysis revealed a strong trend in decreasing risk of epithelial ovarian cancer with increasing cumulative months of pregnancy; this effect was less pronounced in women aged 50-54 than in younger women. In contrast, a marked reduction in risk was associated with ever having breast fed or used oral contraceptives, while the decrease in risk from additional months of either of these exposures was less than that for pregnancy.

Adult↗
PubMed1994

The effects of HIV-2 infection in a rural area of Guinea-Bissau.

OBJECTIVE: To investigate the clinical and immunologic effects, and pattern of mortality associated with HIV-2 infection. SETTING: A rural community in Guinea-Bissau. METHODS: Serologic screening of 2774 subjects aged > 14 years followed by studies of the prevalence of clinical and immunologic abnormalities among 133 subjects with HIV-2 infection and 160 seronegative controls, and surveillance of mortality among all subjects who were screened during a mean of 2 years of follow-up. RESULTS: Generalized lymphadenopathy was the only clinical abnormality significantly associated with HIV-2 infection. Infection was associated with lower CD4 counts and higher beta 2-microglobulin and neopterin levels. During follow-up, 5.5% of infected subjects died compared with 1.8% of the seronegatives (rate ratio adjusted for age and sex, 3.5; 95% confidence interval ((CI), 1.8-6.7). Proportional hazard regression analysis showed that the rate ratio varied with age (P = 0.003) and there was some evidence that the excess of mortality in infected subjects was, in absolute terms, least in the oldest subjects (trend test; P = 0.08). CONCLUSIONS: The findings support previous suggestions that HIV-2 is less pathogenic than HIV-1; the data also suggest that mortality associated with infection may be lower in older subjects.

Adolescent↗
PubMed1994

Risk factors for HIV-1 seroprevalence among drug injectors in the cocaine-using environment of Rio de Janeiro.

To determine risk factors for HIV-1 among drug injectors in Rio de Janeiro, where cocaine is the dominant drug of injection, subjects were recruited using the criteria and interview instrument of the World Health Organization's Cross-National Study of HIV infection and risk behaviour in injecting drug users. HIV antibody test results were derived both from serum tests and from self-reports of previous tests (documented evidence of self-reported seropositivity was required). The analytical sample consists of 123 subjects, recruited both at drug abuse treatment sites and at street locations. Of 27 subjects with both serological and self-reported antibody status data, 20 reported previous negative tests; of these three had positive sera and may have seroconverted. Seven subjects reporting prior positive serostatus all tested positive. For the 123 subjects, seroprevalence was 34%. Independent significant risk factors in multivariate logistic regression with backwards elimination are: years of injection greater than 5; being a male who has had sex with men in the previous 5 years; and not having taken deliberate steps to protect oneself against AIDS. These findings indicate that homosexual/bisexual male drug injectors may be a bridge group through which HIV is entering drug-injecting networks in Rio de Janeiro. Efforts by drug injectors to reduce their risk of infection seem to have protective effects. This underscores the importance of HIV prevention efforts aimed at drug injectors.

Brazil↗
PubMed2002

Texture of cooked potatoes (Solanum tuberosum). 1. Relationships between dry matter content, sensory-perceived texture, and near-infrared spectroscopy.

Properties of fresh potatoes, including dry matter (DM) content, starch content, and near-infrared (NIR) spectra, were determined and related to the sensory-perceived texture of the steam-cooked samples. To quantify these relationships, three potato cultivars, respectively representing a firm cooking potato (cv. Nicola), a mealy cooking potato (cv. Irene), and a cultivar (cv. Bintje) with intermediate cooking properties, were classified on the basis of three size categories. For each size category and cultivar a DM distribution was determined on the basis of the underwater weight of the individual potatoes. Each DM distribution was divided into three subcategories based on low, medium, and high DM contents. This categorization was performed for freshly harvested potatoes and for potatoes stored for 3 and 6 months, respectively. In total, this resulted in 27 DM distributions, of which 16 were non-normally distributed, and 81 samples. Linear relationships were established between the DM content, as determined by either underwater weight analysis or oven-drying, and the starch content. On the basis of partial least-squares regression (PLSR), statistical models were developed relating sensory-based texture descriptors with the DM matter content of the samples. It was also shown, by applying PLSR, that the NIR spectra, originating from the potato samples, could be related to the DM content and to the sensory-perceived texture. From the relationships between DM content and sensory-perceived texture, on the one hand, and from the DM content and NIR spectra, on the other, it was concluded that the DM content rather than the cultivar determines the sensory-perceived texture of steam-cooked potatoes. Cultivar-specific elements may also contribute to the perceived texture but are overruled by the DM content. Storage did not affect the mutual relationships between the DM content, the sensory properties, and the NIR spectra.

Food Technology↗
PubMed1991

Regression methods for estimating attributable risk in population-based case-control studies: a comparison of additive and multiplicative models.

A regression method that utilizes an additive model is proposed for the estimation of attributable risk in case-control studies carried out in defined populations. In contrast to previous multivariate procedures for the estimation of attributable risk, which have utilized logistic regression techniques to adjust for confounding factors, the model assumes an additive relation between the covariates included in the regression equation. As an empirical example, additive and logistic models were fitted to matched case-control data from a population-based study of childhood astrocytoma brain tumors. Although both models fitted the data well, the additive model provided a more satisfactory estimate of the risk attributable to multiple exposures, in the absence of significant additive interaction. In contrast to the results from the logistic model, the adjusted estimates of the risk attributable to each factor included in the additive model summed to the overall estimate for all of the factors considered jointly. Thus, the additive approach provides a useful alternative to existing procedures for the multivariate estimation of attributable risk when the additive model is determined to be appropriate on the basis of goodness-of-fit.

Astrocytoma↗
PubMedDate not supplied

On combining dose-response data from epidemiological studies by meta-analysis.

Using data from a meta-analysis of the effects of oestrogen replacement therapy on the development of breast cancer, we compared alternative methods for combining dose-response slopes from epidemiological studies. We evaluated issues related both to summarizing data from single studies and to combining results from multiple studies. Findings related to the analysis of individual dose-response studies include: (1) a method of weighing studies that gives greater influence to dose-response slopes that conform to the linear relation of relative risk to duration can lead to large differences in calculated weights as a function of non-linearity; (2) a regression model using a variable-intercept resulted in a mean dose-response slope that increased as much as threefold when compared with the values obtained with a zero-intercept model. When combining results from multiple studies, we found: (1) calculating standard errors of mean dose-response slopes by methods that allow for both among-study and within-study variability (a random-effects type model) gave values different from a method that assumes homogeneity and equal within-study precision (a fixed-effects model); (2) the random-effects model gives mean and standard error results most similar to a bootstrap resampling method as increasing heterogeneity is observed (however, this model could give biased mean estimates compared with the bootstrap method); (3) a components-of-variance model compares favourably with the bootstrap and is easier to apply than the random-effects model. Based on these findings, we recommend the use of methods which incorporate heterogeneity to guard against underestimating the standard error. However, caution is urged because bias in point estimates can occur if extreme heterogeneity is present. Two other observations affect the interpretation of data combined from multiple studies. First, inclusion into a model of quality scores assigned by blinded reviewers had little effect on the mean dose-response slope and its standard error. Second, the number of studies required to achieve desired statistical power, varies with effect size.

Analysis of Variance↗
PubMed2001

Risk factors associated with new onset tachyarrhythmias after cardiac surgery--a retrospective analysis.

BACKGROUND: Tachyarrhythmias (TA) represent a frequent and serious problem after cardiac surgery. We retrospectively analyzed 987 cardiac surgery patients admitted to a surgical intensive care unit between 1996 and 1999 to assess incidence and risk factors associated with development of postoperative TA in the intensive care unit. METHODS: TA (n=149) were defined as non-sinus rhythm with a heart rate (HR) > or =100 bpm in patients with preoperative sinus rhythm or as heart rate > or =130 bpm in patients with preoperative atrial fibrillation. A total of 787 patients served as controls (C). Demographic, premorbidity and perioperative data, admission SAPS and MODS-score, presence of clinical syndromes systemic inflammatory response syndrome (SIRS) and sepsis were univariately compared between groups. For prediction of independent risk factors for TA-development two multiple logistic regression models were finally established. RESULTS: Concerning TA, atrial fibrillation and flutter (76%) were observed most frequently, followed by paroxysmal supraventricular tachycardia (15%) and ventricular tachycardia/fibrillation (11%). Age, a history or presence of congestive heart failure, development of SIRS and sepsis, severity of multiple organ dysfunction syndrome and in particular severity of cardiovascular failure proved to be independent risk factors for development of TA. CONCLUSION: In cardiac surgery patients, age, a history or presence of congestive heart failure, postoperative development of a systemic inflammatory response syndrome or sepsis and the severity of multiple organ function syndrome were independent predictors for development of TA in the intensive care unit. The association of severity of cardiovascular dysfunction with TA strongly suggests a causal relationship between catecholamine therapy and TA-development.

Aged↗
PubMed2006

Application Fourier transform near infrared spectrometer in rapid estimation of soluble solids content of intact citrus fruits.

Nondestructive method of measuring soluble solids content (SSC) of citrus fruits was developed using Fourier transform near infrared reflectance (FT-NIR) measurements collected through optics fiber. The models describing the relationship between SSC and the NIR spectra of citrus fruits were developed and evaluated. Different spectra correction algorithms (standard normal variate (SNV), multiplicative signal correction (MSC)) were used in this study. The relationship between laboratory SSC and FT-NIR spectra of citrus fruits was analyzed via principle component regression (PCR) and partial least squares (PLS) regression method. Models based on the different spectral ranges were compared in this research. The first derivative and second derivative were applied to all spectra to reduce the effects of sample size, light scattering, instrument noise, etc. Different baseline correction methods were applied to improve the spectral data quality. Among them the second derivative method after baseline correction produced best noise removing capability and yielded optimal calibration models. A total of 170 NIR spectra were acquired; 135 NIR spectra were used to develop the calibration model; the remaining spectra were used to validate the model. The developed PLS model describing the relationship between SSC and NIR reflectance spectra could predict SSC of 35 samples with correlation coefficient of 0.995 and RMSEP of 0.79 degrees Brix.

Biochemistry↗
PubMed1999

Cardiovascular risk factors as predictors of mortality in type II diabetic patients.

BACKGROUND: Mortality among diabetic patients is high and attempts have been made to find clinical markers which would help to identify patients at risk. MAIN PURPOSE: The purpose of this study was to assess whether cardiovascular risk factors like previous myocardial infarction, left ventricular dysfunction, hypertension, renal insufficiency, and 24-hour mean or minimum heart rate, significantly contribute to the mortality in this group of patients. METHODS: 45 male patients with type II diabetes mellitus were studied by ambulatory electrocardiography to obtain the information about the 24-hour mean and minimum heart rates; records from baseline hospitalization were reviewed for clinical variables. After 55 months mortality data were retrieved. Cox's proportional hazards regression model was used to study the effect of clinical variables on mortality in multivariate analysis. RESULTS: All-cause mortality was 38%. The groups of survivors and nonsurvivors differed significantly in terms of age, the presence of left ventricular dysfunction and renal insufficiency. No significant difference in 24-hour mean or minimum heart rate was registered. In the Cox's model, the presence of left ventricular dysfunction, renal insufficiency and age significantly and independently predicted mortality. CONCLUSION: The present study, showed that left ventricular dysfunction and renal insufficiency were related to the all-cause mortality of type II diabetic patients independently of age, after 55 months of follow-up. Prognostic significance of 24-hour mean or minimum heart rate has not been proven.

Adult↗
PubMedDate not supplied

Ethnic and environmental factors affecting fetal growth in Papua New Guinea.

We studied 738 live singleton births in Papua New Guinea to examine the contribution of ethnic origin and environmental factors to birth weight and gestation. Maternal history, examination and post-partum anthropometry, infant anthropometry and clinical gestational assessment were performed, and placental blood slides and histology were examined. Mothers from Highlands provinces who delivered on the coast had larger babies (mean birthweight 3.34 SD 0.48 kg) than either indigenous coastal mothers (mean 3.10 SD 0.51 kg) or those migrating to Port Moresby from other coastal regions (mean 3.14 SD 0.51 kg). This was due to a combination of longer gestation and better intrauterine growth (assessed by birthweight standard deviation scores). Gestation was slightly longer in Highlands mothers delivering on the coast compared with those delivering in the Highlands, but birthweight standard deviation scores were unchanged. Malarial infection of the placenta was almost never found. Stepwise regression analysis showed that parity, maternal height, Highlands origin and maternal body mass index (weight/height2) were all significant predictors of birthweight standard deviation score, while maternal body mass index and Highlands origin were predictors of gestational length. We conclude that Highland mothers have a tendency to deliver heavier babies, and do not demonstrate the fetal growth-retarding effects of altitude seen in other races. Although the incidence of low birthweight was only 7.1% in our study, we found that indices of maternal nutrition were predictors of birth weight.

Alcohol Drinking↗
PubMed1996

Optimization of selectivity and resolution in micellar electrokinetic capillary chromatography with a mixed micellar system of sodium dodecyl sulfate and sodium cholate.

Selectivity and resolution were studied for the separation of seven corticosteroids by micellar electrokinetic capillary chromatography (MEKC) using a mixed micellar solution of sodium dodecyl sulfate (SDS) and sodium cholate (SC), buffered with 3-(N-morpholino)propanesulfonic acid (MOPS) or 3-[(1,1-dimethyl-2-hydroxyethyl)amino]-2-hydroxypropane sulfonic acid (AMPSO). The changes in selectivity were compared for the AMPSO-SDS-SC system by varying the pH and the concentrations of AMPSO, SDS and SC. The experimental design started with the central composite design and continued in a sequential manner. The optimum selectivity for the separation of the corticosteroids was calculated from the analyte migration times and the analyte velocities, by using empirical quadratic regression models. Satisfactory regression fits and coefficients of determination for prediction were obtained with cross-validated models. To optimize the resolution, the physical parameters of capillary length and analysis time were varied under the conditions optimal for the selectivity. In both the selectivity and the resolution, optimization the overall optimum was determined by using the desirability function technique. Analysis times were controlled by using 1,3-diaminopropane to influence the electroosmotic flow velocity (veo). The voltage was kept constant, which resulted in higher electric field strength in shorter capillaries. No changes in the selectivity were observed when 1,3-diaminopropane was used to control the electroosmotic flow velocity. Such an optimization technique, where the chemical and physical factors affecting the separation are treated independently, seemed to be effective for finding the best possible resolution for the corticosteroids.

Adrenal Cortex Hormones↗