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 1,099 records · Page 61Linked to original sources
PubMed1995

Sociodemographic factors associated with the diagnostic staging of breast cancer in southern Italy.

BACKGROUND: There are marked regional differences in breast cancer mortality rates in Italy, probably linked to factors such as diagnostic delay, therapeutic strategies, and biologic and sociodemographic differences. To investigate a possible link between sociodemographic factors (e.g. age, education, and residence) and delay in the diagnosis of breast cancer, data were evaluated from all such patients from our Institute living in the Campania Region of Southern Italy for 1991-1993. METHODS: Patients were grouped into Tis-T1/N0-N+ versus T2-4/N0-N+ and the variables examined were age (< 40, 41-50, 51-60, > 60 years), education (< or = 5 vs. > 5 school years) and residence (urban vs. rural). An analysis was made using the Pearson's Chi-square test and the multiple logistic regression. RESULTS: Statistically significant differences were found for both residence (P = 0.04) and education level (P = 0.03) in the older than 60 years age group, but only for residence (P = 0.03) in the 51-60 years age group. The risks according to Mantel-Haenszel were 1.28 for education (P = 0.08) and 1.32 for residence in rural municipalities (P = 0.05). The odds ratio for residence in rural municipalities, adjusted by education and by the education-residence interaction, was 2.26 (95% confidence interval [CI], 1.12-4.54) in the 51-60 years age group and 1.74 (95% CI, 1.01-3.00) in the older than 60 years age group. CONCLUSIONS: These data clearly indicate that residents of rural municipalities, as well as poorly educated subjects, are more likely than their respective counter-parts to have a delayed diagnosis of breast cancer.

Adult↗
PubMedDate not supplied

Parametric bootstrap and penalized quasi-likelihood inference in conditional autoregressive models.

This paper discusses a variety of conditional autoregressive (CAR) models for mapping disease rates, beyond the usual first-order intrinsic CAR model. We illustrate the utility and scope of such models for handling different types of data structures. To encourage their routine use for map production at statistical and health agencies, a simple algorithm for fitting such models is presented. This is derived from penalized quasi-likelihood (PQL) inference which uses an analogue of best-linear unbiased estimation for the regional risk ratios and restricted maximum likelihood for the variance components. We offer the practitioner here the use of the parametric bootstrap for inference. It is more reliable than standard maximum likelihood asymptotics for inference purposes since relevant hypotheses for the mapping of rates lie on the boundary of the parameter space. We illustrate the parametric bootstrap test of the practically relevant and important simplifying hypothesis that there is no spatial autocorrelation. Although the parametric bootstrap requires computational effort, it is straightforward to implement and offers a wealth of information relating to the estimators and their properties. The proposed methodology is illustrated by analysing infant mortality in the province of British Columbia in Canada.

Algorithms↗
PubMed2001

Modeling the effect of temperature and pH on activity of enzymes: the case of phytases.

In this study, the behavior of enzyme activity as a function of pH and temperature is modeled on the basis of fundamental considerations. A formulation is developed that includes the activation of enzymes with increasing temperatures and the deactivation of enzymes at higher temperature, together with the effect of protonation and hydroxylation on activity at various constant pH levels. The model is calibrated and validated against an extensive set of experimental data on phytases from seven different origins. The percentage variance accounted for (R(2)(adj)), obtained by statistical nonlinear regression analysis on all data sets, was shown to range from 97.6% to 99.5%. The equilibrium constant of protonation and hydroxylation proved to be independent of temperature.

6-Phytase↗
PubMed2000

A general test of association for complex diseases with variable age of onset.

Analysis of age of onset is a key factor in linkage and association studies of some complex genetic traits. Recent methodological developments are mainly concentrated on binary or quantitative traits, in which age of onset information is ignored. We propose a linkage disequilibrium-based Cox model and a robust score test for testing association between a marker and a disease with variable age of onset. The proposed model is semi-parametric, with an unspecified baseline hazard function. Simulation results indicate that the proposed methods have correct error rates and good statistical power, even in the presence of population admixture. This approach offers a solution to the problem of testing for marker associations when there is a variable age of onset.

Age of Onset↗
PubMed2003

Cause, type, and workers' compensation costs of injury to fire fighters.

BACKGROUND: Work-related injury rates in the fire service industry exceed those for most other industries, however little is known about the cost of injury to firefighters. METHODS: This is a preliminary investigation of detailed worker's compensation records of firefighter injuries, 1,343 claims in all, collected from 1992-1999. Summary statistics and regression analysis regarding the cause, nature, and cost of injury to firefighters are presented. RESULTS: Overexertion accounted for a significant portion (over 1/3) of injuries to firefighters, typically involved injuries to the back, and was associated with significantly higher costs than other types of injuries. The per-claim average worker's compensation cost of injury to firefighters was $5,168 and the average for injuries caused by overexertion was $9,715. CONCLUSIONS: Overexertion is a costly source of injury to firefighters that can likely be reduced through policy intervention.

Accidents, Occupational↗
PubMed1988

The use of geometric and arithmetic mean exposures in occupational epidemiology.

In constructing quantitative measures of exposure for the study of chronic occupational disease, researchers have generally used a cumulative exposure calculated as the sum of the products of period-specific exposure concentrations and the time each individual spent in each exposure category. There has been some disagreement and lack of clarity about the use of the geometric or arithmetic mean of exposure for this calculation. This paper explores the difference in the use of the two measures and defines a relative bias introduced with the geometric vs. the arithmetic mean. The magnitude of the bias is calculated in two linear models of possible exposure-response relationships. The theoretical basis for the choice of one mean over the other is then explored. It is suggested that when adopting a linear exposure response model, the arithmetic mean is the more appropriate measure. In other models, such as a linear-log (outcome is proportional to the logarithm of exposure) model, the geometric mean would be more appropriate.

Chronic Disease↗
PubMed2004

Expanded RED products and loci containing CAG/CTG repeats on chromosome 17 (ERDA1) and chromosome 18 (CTG18.1) in trans-generational pairs with bipolar affective disorder.

The purpose of the present study was to further test if expanded CAG repeats detected by the repeat expansion detection (RED) method in bipolar affective disorder (BPAD) are correlated with ERDA1 (17q21.3) and/or CTG18.1 (18q21.1) loci expansions, and changes of phenotype severity in successive generations (anticipation). The sample was designed to analyze ERDA1 and CTG18.1 expansions in trans-generational pairs of affected individuals (parent-offspring pairs: G1 and G2). Clinical and genetic information was available on 95 two-generations pairs. We found in our sample no one patient carrying an expanded allele at the CTG18.1 locus. This observation is true for all individuals in G1 and G2. Using the conditional logistic regression, no statistical difference was observed between the two generations for ERDA1 alleles (chi(2) = 0.2, P = 0.65). These data do not support the correlation between expanded RED products (RED fragments >120) and expanded alleles at ERDA1 in trans-generational pairs with BPAD. We were not able to detect any correlation for CTG18.1. Earlier age at onset in offspring generation was also not associated with expanded RED products explained by expanded ERDA1 alleles.

Adult↗
PubMed1984

Estimation of body fat in female rhesus monkeys.

Measurement of height (crown-rump length), body weight, and abdominal subcutaneous fat depth, based on skinfold thickness taken from 13 female rhesus monkeys comprising two age groups were correlated with body fat values derived from tritiated water determinations of total body water. The manner with which each measure was related to percent body fat differed as a function of age of the animal. In the young, nulliparous females, crown-rump length was the single best predictor of body fat, whereas in the older, multiparous females, skinfold thickness correlated most highly with body fat. When all measurements, including the Quetelet index [(wt/ht)2 x 1,000], were combined statistically and regressed against percent body fat, a significant increase in predictive ability was obtained. When each age group was considered separately, the resulting equations again reflected the age-group biases. In addition, as an internal check on the validity of the regression equations, an additional regression analysis was performed using morphometric data from selected animals in each age group. These equations yielded accurate estimates of body fat when compared to determinations made from total body water. These analyses indicate that the predictive accuracy of morphometric data is greatly enhanced by using these measurements in concert. Furthermore, the utility of such predictions is influenced by the specific physical characteristics of the subject population.

Adipose Tissue↗
PubMed2005

Early seizures after temporal lobectomy predict subsequent seizure recurrence.

Patients are understandably anxious if seizures occur immediately after temporal lobectomy. Such "neighborhood" seizures are commonly regarded as irrelevant to seizure outcome and discounted in outcome measurement. We conducted an in-depth examination of early postoperative seizures (<28 days) and outcome. The risk of recurrence at one postoperative year was calculated using Poisson regression, and statistical adjustments were made for preoperative pathology. Of 321 patients, 69 (22%) experienced early postoperative seizures. These early seizures were associated with subsequent seizure recurrence (rate ratio [RR] 5.9; 95% confidence interval [CI], 4.1-8.4). Among patients with early seizures, the only significant factor was the presence of seizure precipitants, which was associated with a lower recurrence risk. However, when compared with patients with no early seizures, those with precipitants to early seizures had a higher risk of recurrence (RR, 3.0; 95% CI, 1.8-5.2). The risk was higher again for patients without precipitants to early seizures (RR, 7.6; 95% CI, 5.0-11.5). Early seizures and other seizure recurrences in the first postoperative year did not differ in their effect on subsequent outcome (X(2) [3] = 3.4, p = 0.33). We conclude that early postoperative seizures are associated with subsequent seizure recurrence. These findings have implications for patient counseling and the measurement of outcome.

Anterior Temporal Lobectomy↗
PubMed1986

Prognostic factors in breast cancer and the development of a prognostic index.

A number of different factors are known to be correlated with survival of patients with breast cancer. Among these are lymph node status, tumour size, oestrogen receptor (ER), progesterone receptor (PR) and androgen receptor (AR) status. The purpose of this study was to investigate the relative significance of these factors and use this information to construct a prognostic index capable of predicting survival. These factors together with age and menopausal status were studied and correlated with prognosis in 796 women with primary breast cancer. The data were analysed in a stepwise manner by the Cox proportional hazards regression technique. Statistically, greater than 3 nodes involved gave the worst prognosis (P less than 0.001). This was followed by ER if less than 10 fmol/mg cytosol protein (P less than 0.001), PR if less than 10 fmol (P less than 0.01), greater than 0 lymph nodes involved (P less than 0.01) and the number of years over age 65 (P less than 0.01). When these factors were accounted for, tumour size, menopausal status and AR did not significantly improve prediction of survival. The significant factors were incorporated into a prognostic index: I = N + E + P + A, where N = 0 if no nodes involved, 13 (if 1-3 nodes involved and 31 if greater than 3 nodes involved, E = 15 if ER less than 10 fmol, P = 12.5 if PR less than 10 fmol and A = number of years over 65. Using this index five year survival curves were constructed corresponding to groups of patients with widely differing prognoses. Predicted five year survival ranged from 96 to 12 per cent.

Adult↗
PubMed2004

Death penalty support for special offender populations of legally convicted murderers: juveniles, the mentally retarded, and the mentally incompetent.

The U.S. Supreme Court recently re-examined the constitutionality of the death penalty in the context of two of three special offender populations of murderers (juveniles, mentally retarded, and mentally incompetent). The Court reaffirmed the imposition of the death penalty for juveniles 16 and 17, while reversing itself on the mentally retarded. In reaching its decision, the Court relied on society's "evolving standards of decency." Using Likert-type items, this study is the first to have prospective jurors assess support for the death penalty for these specific offender groups. The public's support for the execution of each of the groups is then compared with existing case law. Descriptive statistics and regression analyses indicate that, as expected, the levels of support for the applicability of capital punishment to the various special offender populations are much lower than that for the general adult offender. Moreover, these findings are congruent with the holdings of the Court with one notable exception: a slight majority of respondents supported executing the mentally incompetent. Reasons for the public's apparent departure from the Supreme Court holding prohibiting the execution of mentally incompetent convicted murderers are discussed. The Court's continued role in protecting marginalized populations from "cruel and unusual punishment" is explored in the context of strong public sentiment demanding justice and finality despite changes in offenders' mental capacity.

Adolescent↗
PubMed1994

Sources of variability of human plasma apolipoprotein A-IV levels and relationships with lipid metabolism.

Plasma apolipoprotein (apo) A-IV concentration was determined by immunoelectrophoretic assay (EIA) in 119 nuclear families. No significant effect of concomitants such as age, weight, height, body mass index, tobacco, and alcohol consumption was observed on apo A-IV levels in men and in boys. In women, contraceptive use and hormonal status affected apo A-IV levels. In girls, only age influenced the quantitative phenotype. After adjusting by specific concomitants significant correlations were observed between apo A-IV levels and triglycerides, apolipoprotein A-I and apo B levels, suggesting a role of apolipoprotein A-IV in the hepatic lipid metabolism. Intrafamilial correlations were estimated to investigate the plausibility of a common family factor. The results obtained in this study showed a significant correlation between family members with the exception of mother-daughter pairs. Using a variance components model, the contribution of genetic and environmental factors was then investigated. Different statistical models were used and two major hypotheses were statistically acceptable: the first hypothesis supports that shared and specific environmental factors explain 35 and 65%, respectively, of the total adjusted plasma apo A-IV variation. The fraction of apo A-IV variability attributable to genetic factors was null. The second hypothesis supports that the fraction of variability attributable to apo A-IV genetic variation is 67% and the common spouse environmental factors are responsible for 33% of the total variability and no specific environmental effect was found. Among the two hypotheses, taking account of the metabolism function, we support the first one without excluding gene-environment interactions which could mask the genetic influence.

Adult↗
PubMed1990

Injectable contraceptives and risk of invasive cervical cancer: evidence of an association.

In a case-control study conducted in Latin America, the relationship of injectable contraceptive (IC) use to risk of invasive cervical cancer was analyzed while controlling for a variety of other risk factors, including female and spouse sexual behavior and infection with human papillomaviruses (HPV). Thirty-two cases and 82 controls reported ever having used IC. Women reporting use of IC for less than 5 years had an adjusted RR of 0.5 (95% Cl = 0.3-0.9), but users for 5 or more years had an RR of 2.4 (95% Cl = 1.0-5.7). The effect of prolonged IC use was stronger for women reporting first use 10 or more years before interview (adjusted RR = 3.4, 95% Cl = 1.1-24.9) and more than 5 years since last use (adjusted RR = 5.3, 95% Cl = 1.1-10.0). Cervical cancer risk associated with prolonged IC use was particularly high among women who reported never having had a Pap smear or having had one 2 or more years before interview (adjusted RR = 6.3, 95% Cl = 2.1-18.7). The reduced cervical cancer risk associated with short-term use of IC may reflect intensive Pap smear screening as the method is initiated. Although hampered by small numbers, these results suggest an adverse effect of prolonged IC use on cervical cancer risk, particularly among women who cease participation in screening programs after terminating usage, and indicate that long-term IC users should be monitored for cervical disease until more conclusive results are available.

Case-Control Studies↗
PubMed2006

Evaluation of the protonation thermochemistry obtained by the extended kinetic method.

An evaluation of the results obtained by the extended kinetic method for a series of representative bases is presented here. Analysis of the original experimental data is conducted using the orthogonal distance regression (ODR) statistical treatment. A comparison with the proton affinities and protonation entropies obtained from variable temperature equilibrium constant measurements demonstrate deviations, which may be ascribed to random and systematic errors. Considerable random errors are associated with the extended kinetic method if the number of reference bases and the range of effective temperatures are too low. It is also confirmed that large systematic errors on proton affinities and protonation entropies are obtained when large protonation entropy is associated with the considered system. It is, however, encouraging to note that the gas phase basicities obtained by the extended kinetic method are generally comparable to that obtained by other methods within a few kJ/mol.

Journal Article↗
PubMed2003

Steady-state flux and lag time in the stratum corneum lipid pathway: results from finite element models.

Finite element model (FEM) solutions of the diffusion through two-dimensional representations of the stratum corneum (SC) lipid pathway are presented. Both simplified, regular "brick and mortar" models and a more complex, irregular model are analyzed. It is assumed that diffusion occurs only within the SC lipids and the lipids are isotropic. The steady-state flux and lag time are solved and compared with the corresponding values for a homogeneous membrane of the same thickness consisting of lipid material. Results confirm that the heterogeneous SC model behaves like a homogeneous membrane, meaning that FEM diffusion simulations are well approximated by an appropriate solution of the diffusion equation for a homogeneous membrane. Additionally, both steady-state flux and lag time (relative to these values in a homogeneous membrane) can be predicted from algebraic equations based on simple dimensionless descriptors of SC geometry. However, values for diffusivity derived from homogeneous membrane approximations to the FEM solutions (effective diffusivity, D*) are not equal to the intrinsic diffusivity of the chemical in lipid. Furthermore, the pathlength derived from homogeneous membrane approximations to FEM solutions (effective pathlength, l*) is not equal to the lipid pathlength and is not dependent on SC tortuosity. Whereas l* is not a function of corneocyte overlap, D* is. These model results suggest that diffusion properties of the SC lipid pathway can be correlated to SC geometry, but intrinsic diffusion coefficients and SC tortuosity cannot be derived from common diffusion cell experiments. Use of the model equations to predict permeability and lag time of lipophilic solutes is described.

Algorithms↗
PubMed2005

Mixed models incorporating intra-familial correlation through spatial autoregression.

Family researchers are challenged by the need to account for the special forms of statistical dependence that can exist in family data. To address this issue, mixed modeling methods were adapted to account for dependence of continuous outcomes measured across multiple family members. This was accomplished using a spatial autoregressive approach that accounts for dependence on direction as well as on distance apart. For family data, the dimensions underlying direction can correspond to different family members, thereby accounting for different correlations between family members. When the data are also longitudinal, a dimension representing distance apart in time also can be included to account for temporal correlation. Fixed effects involving general linear models can be included as well. Example analyses were conducted to demonstrate the use of the spatial autoregressive approach for modeling intra-familial correlation.

Clinical Nursing Research↗
PubMed2004

How well do patients report noncompliance with antihypertensive medications?: a comparison of self-report versus filled prescriptions.

PURPOSE: To address poor patient compliance with antihypertensives, clinicians and researchers need accurate measures of adherence with prescribed regimens. Although self-reports are often the only means available in routine practice, their accuracy and agreement with other data sources remain questionable. METHODS: A telephone survey was conducted on 200 hypertensive patients treated with a single antihypertensive agent in a large health maintenance organization (HMO) or a Veterans Affairs medical center (VAMC) to obtain self-reports of the frequency of missing antihypertensive therapy. We then analyzed records of all filled prescriptions to calculate the number of days that patients actually had antihypertensive medications available for use. Agreement between the two data sources was measured with correlation coefficients and kappa statistics. Logistic regression models were used to identify demographic, clinical and psychosocial correlates of overstating compliance. RESULTS: There was very poor agreement between self-reported compliance and days actually covered by filled prescriptions (Spearman correlation coefficient = 0.15; 95%CI: 0.01, 0.28). Very poor agreement was also observed between a categorical measure of self-reported compliance (ever vs. never missing a dose) and categories of actual compliance defined by filled prescriptions (< 80% vs > 80% of days covered; kappa = 0.12, 95%CI: -0.02, 0.26). Surprisingly, few factors were associated with inaccurate self-reporting in either crude or adjusted analyses; fewer visits to health care providers was significantly associated with overstating compliance. CONCLUSIONS: Compliance was markedly overstated in this sample of patients and few characteristics identified those who reported more versus less accurately. Clinicians and researchers who rely on self-reports should be aware of these limits and should take steps to enhance their accuracy.

Aged↗