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 883 records · Page 49Linked to original sources
PubMed1994

Hierarchical regression for epidemiologic analyses of multiple exposures.

Many epidemiologic investigations are designed to study the effects of multiple exposures. Most of these studies are analyzed either by fitting a risk-regression model with all exposures forced in the model, or by using a preliminary-testing algorithm, such as stepwise regression, to produce a smaller model. Research indicates that hierarchical modeling methods can outperform these conventional approaches. These methods are reviewed and compared to two hierarchical methods, empirical-Bayes regression and a variant here called "semi-Bayes" regression, to full-model maximum likelihood and to model reduction by preliminary testing. The performance of the methods in a problem of predicting neonatal-mortality rates are compared. Based on the literature to date, it is suggested that hierarchical methods should become part of the standard approaches to multiple-exposure studies.

Algorithms↗
PubMed1990

Undernutrition among Bedouin Arab infants: the Bedouin Infant Feeding Study.

Two hundred seventy-four healthy Bedouin Arab newborns in 1981 were followed for 18 mo to examine the relationship between infant-feeding practices and growth during planned social change. Although wasting was not prevalent, the prevalence rate of stunting (less than or equal to -2 SDs) increased from 12% to 19% to 32% at 6, 12, and 18 mo, respectively. After multiple-logistic-regression adjustment for covariates, the odds ratio (OR) of stunting at 6 mo was reduced among infants breast-fed only or fed with supplement compared with weaned infants. Infant-feeding practices were not associated with stunting in later infancy; however, those stunted at 6 mo had an OR of 13 of stunting at 12 mo and those stunted at 12 mo had an OR of 14 of stunting at 18 mo. In a multiple-linear-regression analysis, seasonality, duration of breast-feeding, hospitalized morbidity, and residual of height at 6 mo were negatively associated with daily average linear growth from 6 to 12 mo; these factors only explained 12% of the variation in daily linear growth.

Breast Feeding↗
PubMed1996

Simulation study of hierarchical regression.

Hierarchical regression - which attempts to improve standard regression estimates by adding a second-stage 'prior' regression to an ordinary model - provides a practical approach to evaluating multiple exposures. We present here a simulation study of logistic regression in which we compare hierarchical regression fitted by a two-stage procedure to ordinary maximum likelihood. The simulations were based on case-control data on diet and breast cancer, where the hierarchical model uses a second-stage regression to pull conventional dietary-item estimates toward each other when they have similar levels of food constituents. Our results indicate that hierarchical modelling of continuous covariates offers worthwhile improvement over ordinary maximum-likelihood, provided one does not underspecify the second-stage standard deviations.

Bayes Theorem↗
PubMed2005

Development of a global stochastic model relating the distribution of individual cell and population physiological states.

Our ability to predict the lag (lambda) prior to growth of foodborne pathogens is limited by our lack of understanding of the physiological changes taking place in the individual cell during the adaptation process. Theoretical models have been developed to describe the stochastic nature of individual cells, and probability distributions have been used to assign hypothetical values of the physiological state to individual cells (p(i)). The aim of this study is to develop a polynomial model which will link distributions of p(i) values to the physiological state of the population (h(0)), and thus to the lambda. Risk analysis software was used to simulate values of p(i) for populations of cells drawn from lognormal distributions with parameters alpha and beta, and growth curves were simulated using a modified continuous-discrete-continuous (CDC) model. Values for h(0) were then obtained for each growth curve by fitting with the heterogeneous population model (HPM). Multiple regression analysis was used to develop a polynomial function which described the subsequent h(0) value as a function of alpha and beta (R(2)=0.9957). Outputs from simulations using the polynomial model agree well with results from related stochastic models, and suggest that distributions can accurately describe the physiological state of cell populations.

Adaptation, Physiological↗
PubMed1989

Utilisation of acute care hospital beds by levels of care.

An exploratory analysis of utilisation patterns of acute care hospitals in the Province of Alberta, Canada was carried out to develop a methodology for assessing bed utilisation profiles of acute care hospitals by levels of care. The utilisation of Alberta acute care hospital beds was measured in terms of primary, secondary and tertiary levels of hospital services. Patient origin-destination methodology was applied and a regionalisation perspective employed. The data used for this study were hospital separation abstracts compiled by all Alberta acute care hospitals during year 1986, this coincided with the most recent available Canadian census data. It was estimated that approximately 10-11% of Alberta beds were used for tertiary care as derived from population based utilisation rates and patient flow patterns. With respect to per capita measurement, the number of beds used per 1,000 residents was: 3.5 to 3.9 for primary, 1.2 to 1.6 for secondary, and about 0.6 for tertiary levels of care. Regression analysis revealed that the marginal cost per bed at each level was approximately 75-79, 87-88, and 201-209 thousand Canadian dollars per year in 1986 for primary, secondary and tertiary care respectively. The profiles thus estimated explained about 65% of per bed hospital cost variation.

Alberta↗
PubMed1995

Modeling household fertility decisions: estimation and testing of censored regression models for count data.

"This paper adds to the recent body of research on fertility by estimating and testing censored Poisson regression models and censored negative binomial regression models of household fertility decisions. A novel feature of this study is that in each case the censoring threshold varies from individual to individual. Also, a Lagrange multiplier or score test is used to investigate overdispersion. In these regression models the dependent variable is the number of children. In this situation, censored Poisson regression models and censored negative binomial regression models have statistical advantages over OLS, uncensored Poisson regression models, and uncensored negative binomial regression models. The censored models employed in this study are estimated using panel data collected from the Consumer Expenditure Survey compiled by the [U.S.] Bureau of Labor Statistics."

Americas↗
PubMed1987

Preschool child feeding, health and nutritional status in Gualaceo, Ecuador.

Anthropologic, survey, dietary recall and anthropometric techniques were used to study the correlates of nutritional status of preschool children under five years of age in Gualaceo Ecuador. A widespread stunting was found among the children. Nutritional status was worst among infants comprised between 12 and 23 months old but it improved between April and August, thus suggesting seasonality changes of nutritional status. The correlates of nutritional status (expressed as Z score of weight-for-age) were dietary diversity, birth-spacing, fertility, migration, household income, material goods owned, and parental education. A regression model with these variables predicted 63% of the variability in weight-for-age. The prediction of height-for-age was similar, but only predicted 43% of the variability in height-for-age. Correlates of dietary diversity, birth-spacing, fertility, and migration were child age, maternal age and arm circumference, parental education, use of birth control, household food expenditure, material goods owned, and the raising of animals. Parental education was a correlate of dietary diversity, fertility and migration. Parental education was related to change in weight-for-age in the longitudinal subset. Pre-harvest time and a pathway of illness leading to decreased dietary diversity and to decreased nutritional status in April, were suggested as important to preschool child nutritional status. Hot-cold ideology--resulting in food withdrawal during illness and restriction of high-protein and high-calorie foods--appears to be an important mechanism determining preschool child nutritional status. Breast-feeding, sanitary, higienic, birth control, and drinking (alcohol) practices were suggested as areas that could be improved, in order to improve preschool child nutritional status. Communication between parents and western health care providers was also suggested as an area for improvement.

Body Height↗
PubMed1988

Multifactor prediction of hospital service workload.

Prediction of workload for hospital departments is a forecasting problem. Most forecasting techniques use time series methods. However, the unpredictability of hospital utilization which occurs because of environmental turbulence requires that predictions be based on strategic plans. Regression models developed from cross-sectional data can provide a methodology for linking predictions of service workload to strategic plans. The methodology is illustrated and evaluated in a pilot study, which produced models predicting measures of workload such as laboratory tests (R2 = .85), doses of medication given to inpatients (R2 = .43), outpatient prescriptions (R2 = .79), and x-rays (R2 = .86).

Forecasting↗
PubMed2004

A nonparametric approach to calculate critical micelle concentrations: the local polynomial regression method.

The application of a statistical method, the local polynomial regression method, (LPRM), based on a nonparametric estimation of the regression function to determine the critical micelle concentration (cmc) is presented. The method is extremely flexible because it does not impose any parametric model on the subjacent structure of the data but rather allows the data to speak for themselves. Good concordance of cmc values with those obtained by other methods was found for systems in which the variation of a measured physical property with concentration showed an abrupt change. When this variation was slow, discrepancies between the values obtained by LPRM and others methods were found.

Algorithms↗
PubMed1998

Estimating the effect of zidovudine on Kaposi's sarcoma from observational data using a rank preserving structural failure-time model.

Researchers commonly express scepticism about using observational data to estimate the effect of a treatment on an outcome the treatment is intended to affect. In this paper, we consider using data from the Multicenter AIDS Cohort Study (MACS) to determine whether zidovudine prevents the development of Kaposi's sarcoma among HIV-positive gay men. Several methodologic issues common to observational data characterized the study: information on potentially important confounders was missing at some study visits; investigators did not always know the time of changes in treatment level, nor the value of confounders at that time, and the censoring process depended strongly on time-varying covariates related to outcome. We describe application to our data of Robins' paradigm for defining, modelling and estimating the effect of a time-varying treatment and show how to modify his approach to deal with the methodologic issues we have mentioned. Further, we demonstrate that relative risk regression is less well equipped to deal with these issues. We compare our results to the findings from randomized trials, and conclude that observational studies may sometimes be useful in evaluating the effect of treatment on an intended outcome.

Acquired Immunodeficiency Syndrome↗
PubMed2004

The amygdala processes the emotional significance of facial expressions: an fMRI investigation using the interaction between expression and face direction.

Neuroimaging studies have shown activity in the amygdala in response to facial expressions of emotion, but the specific role of the amygdala remains unknown. We hypothesized that the amygdala is involved in emotional but not basic sensory processing for facial expressions. To test this hypothesis, we manipulated the face directions of emotional expressions in the unilateral visual fields; this manipulation made it possible to alter the emotional significance of the facial expression for the observer without affecting the physical features of the expression. We presented angry/neutral expressions looking toward/away from the subject and depicted brain activity using fMRI. After the image acquisitions, the subject's experience of negative emotion when perceiving each stimulus was also investigated. The left amygdala showed the interaction between emotional expression and face direction, indicating higher activity for angry expressions looking toward the subjects than angry expressions looking away from them. The experienced emotion showed the corresponding interaction. Regression analysis showed a positive relation between the left amygdala activity and experienced emotion. These results suggest that the amygdala is involved in emotional but not visuoperceptual processing for emotional facial expressions, which specifically includes the decoding of emotional significance and elicitation of one's own emotions corresponding to that significance.

Affect↗
PubMed1999

Monitoring and assessing symptoms of chronic fatigue syndrome: use of time series regression.

Chronic Fatigue Syndrome's principal symptoms are severe and include prolonged fatigue and a number of other minor symptoms. Behavioral data collection methods were used in a case study to show some of the benefits that can be derived from monitoring symptoms hourly and daily. Using time series regression, several statistically significant correlates of fatigue were found both within days and between days. Perceived energy, physical exertion, and mental exertion were significantly related to fatigue in both analyses. Collection of such data may help resolve a number of theoretical and methodological problems in research on the Chronic Fatigue Syndrome.

Adult↗
PubMed1991

Civil war and the spread of AIDS in Central Africa.

Using ordinary least squares regression techniques this paper demonstrates, for the first time, that the classic association of war and disease substantially accounts for the presently observed geographical distribution of reported clinical AIDS cases in Uganda. Both the spread of HIV 1 infection in the 1980s, and the subsequent development of AIDS to its 1990 spatial pattern, are shown to be significantly and positively correlated with ethnic patterns of recruitment into the Ugandan National Liberation Army (UNLA) after the overthrow of Idi Amin some 10 years earlier in 1979. This correlation reflects the estimated mean incubation period of 8-10 years for HIV 1 and underlines the need for cognizance of historical factors which may have influenced current patterns of AIDS seen in Central Africa. The findings may have important implications for AIDS forecasting and control in African countries which have recently experienced war. The results are compared with parallel analyses of other HIV hypotheses advanced to account for the reported geographical distribution of AIDS in Uganda.

Acquired Immunodeficiency Syndrome↗
PubMedDate not supplied

High-risk sexual behavior in the general population. Results from a national survey, 1988-1990.

The responses of 2,896 adults who completed the General Social Survey (1988-1990), a nationally representative household probability sample of the United States adult population, were analyzed. Three outcome variables were examined: engaging in sexual intercourse with two or more partners, with five or more partners, or with a stranger in the past year. Age, marital status, gender, pattern of alcohol consumption, and race have the strongest and most consistent relationship with having multiple sexual partners or sex with a stranger. Marriage reduces the odds of having 5 or more sexual partners by a factor of 90% (odds ratio, OR, = 0.10). For each single year increase in age, the odds of having multiple partners or sex with a stranger also decrease (OR = 0.95). Alcohol consumption, on the other hand, increases the odds of sexual risk behavior by a factor of 2 to 3 in the three models. Men are more likely to have 5 or more sexual partners (OR = 7.17) and sex with a stranger (OR = 5.62) than women; and blacks are more likely to have multiple partners (OR = 2.82) than members of other racial or ethnic groups. In the United States last year, an estimated 3 to 6 million adults had sex with 5 or more partners and an estimated 5 to 8 million had sex with a stranger.

Adolescent↗
PubMed1998

Generalized additive models for current status data.

Current status data arise in studies where the target measurement is the time of occurrence of some event, but observations are limited to indicators of whether or not the event has occurred at the time the sample is collected--only the current status of each individual with respect to event occurrence is observed. Examples of such data arise in several fields, including demography, epidemiology, econometrics and bioassay. Although estimation of the marginal distribution of times of event occurrence is well understood, techniques for incorporating covariate information are not well developed. This paper proposes a semiparametric approach to estimation for regression models of current status data, using techniques from generalized additive modeling and isotonic regression. This procedure provides simultaneous estimates of the baseline distribution of event times and covariate effects. No parametric assumptions about the form of the baseline distribution are required. The results are illustrated using data from a demographic survey of breastfeeding practices in developing countries, and from an epidemiological study of heterosexual Human Immunodeficiency Virus (HIV) transmission.

Adult↗
PubMed1996

Using the theory of reasoned action (TRA) to understand the decision to use condoms in an STD clinic population.

The theory of reasoned action (TRA) provides useful information when designing health education interventions. In this study, 703 heterosexual STD clinic clients responded to a TRA-based survey. With steady partners, social norms and attitudes toward condom use were significant predictors of intention for both men and women. The interaction of attitude and norm increased prediction for men (R = .64, p < 0.001) and women (R = .70, p < 0.001). With casual partners, attitude was a predictor for men and social norm was a predictor for women. Prior use of condoms increased prediction for men (R = .38, p < 0.001) and women (R = .47, p < 0.001). Findings suggest that, in addition to traditional TRA model variables, the relationship between sexual partners and the individual's prior experience with condom use should be incorporated into attempts to understand this complex, dyadic behavior. Examining specific outcome and normative beliefs also provides important information for intervention design.

Adult↗
PubMed1991

Social class as a risk factor for infant mortality in an Australian population.

Studies in other countries have identified social class as a risk factor for infant mortality. In Australia there is no systematic collection of population data by social class, partly due to the absence of a recognized measure. The use of occupational prestige as an indicator of social class is discussed and Australian prestige scales reviewed. In a population based study, logistic regression analysis of infant mortality in an Australian (NSW) population shows the effects of social class on infant mortality which remain when maternal age, marital status and parity are controlled.

Australia↗
PubMed1994

Factors affecting number of prenatal care visits during second pregnancy among adolescents having rapid repeat births.

PURPOSE: To examine factors associated with the number of prenatal care visits during second pregnancy for adolescents having a short interval between pregnancies. METHODS: The sample includes all adolescents aged 13 to 17 years whose first pregnancy resulted in a birth at a regional medical center in southeastern North Carolina from January 1983 to December 1989 and who had a repeat pregnancy within 24 months which resulted in a birth. We abstracted data from medical records and birth certificates. We fit a negative binomial regression model to determine the effects of various factors on the number of prenatal care visits during second pregnancy. RESULTS: The number of prenatal care visits during the first pregnancy, poor first birth outcome, interval between first and second pregnancy, and care provided by health department staff during first pregnancy were all positively associated with number of prenatal care visits during second pregnancy when controlling for gestation age of second birth. Other independent variables in the model included maternal age, education, black race, and being unmarried at the time of second birth. CONCLUSIONS: Because prenatal care is important for healthy mothers and babies, adolescents should be encouraged to seek prenatal care early in the first pregnancy. This could be an important time to implement interventions aimed at increasing prenatal care utilization in this and subsequent pregnancies.

Adolescent↗