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 523 records · Page 29Linked to original sources
PubMed2001

Clinical and demographic prognostic indicators for human disc cell proliferation in vitro: pilot study.

STUDY DESIGN: Human anulus cells were cultured under control and experimental conditions to study associations between proliferation and clinical-demographic features of subjects from which cells were obtained. Statistical multiple regression analyses were applied to develop mathematic models relating proliferation to age, gender, Thompson score (denoting stage of disc degeneration), and status (control donor [postmortem]; surgical patient). OBJECTIVES: To identify the effect of donor characteristics on proliferative capacities of human disc cells. SUMMARY OF BACKGROUND DATA: As therapeutic options for disc degeneration increase, novel biologic options are important future considerations. Little is known about the influence of clinical-demographic features on cell proliferation. METHODS: Anulus cells were studied in two designs: 1) Cells from 12 individuals were grown in monolayer with 50 ng/mL interleukin growth factor-1 (IGF-I), 100 ng/mL insulin, or control conditions. 2) Cells from nine individuals were grown in three-dimensional culture with 10 ng/mL IGF-I or control conditions. Cell proliferation data and data on age, gender, Thompson score, and status were collected. Standard statistical analyses were used to develop correlation models. RESULTS: Data from monolayer experiments produced significant models fitting proliferation in the presence of low serum, 50 ng/mL IGF-I, or insulin, with age, gender, Thompson score, and status (respective R2: 0.827, 0.680, 0.850). Three-dimensional cultures exposed to 10 ng/mL IGF-I resulted in proliferation that correlated in a significant negative manner with Thompson score (r = -0.798). CONCLUSIONS: Clinical-demographic prognostic indicators may help predict levels of proliferation. Greater age, greater disc degeneration, female gender, and surgical derivation had deleterious effects on proliferation potential in this model.

Adolescent↗
PubMed1989

Female to male transmission of human immunodeficiency virus type 1: risk factors for seroconversion in men.

To determine the frequency and risk factors for female to male sexual transmission of human immunodeficiency virus type 1 (HIV-1), a prospective study was carried out in 422 men who had acquired a sexually transmitted disease (STD) from a group of prostitutes with a prevalence of HIV-1 infection of 85%. The initial seroprevalence of HIV among the men was 12%. 24 of 293 (8.2%) initially seronegative men seroconverted to HIV-1. Newly acquired infection was independently associated with frequent prostitute contact (risk ratio 3.2, 95% confidence interval 1.2-8.1), with the acquisition of genital ulcer disease (risk ratio 4.7, 95% confidence interval 1.3-17.0), and with being uncircumcised (risk ratio 8.2, 95% confidence interval 3.0-23.0). 96% of documented seroconversions occurred in men with one or both of the latter two risk factors. In a subgroup of 73 seronegative men who reported a single prostitute sexual contact, the frequency of HIV-1 infection was 8.2% during 12 weeks of observation. No man without a genital ulcer seroconverted. A cumulative 43% of uncircumcised men who acquired an ulcer seroconverted to HIV-1 after a single sexual exposure. These data indicate an extremely high rate of female to male transmission of HIV-1 in the presence of STD and confirm a causal relation between lack of male circumcision, genital ulcer disease, and susceptibility to HIV-1 infection.

Acquired Immunodeficiency Syndrome↗
PubMed1988

Clearance of levonorgestrel from the circulation following removal of NORPLANT subdermal implants.

The disappearance of levonorgestrel from plasma after the removal of NORPLANT subdermal implants was studied in 12 women who had been treated from 5.5 to 78 months. The existence of one or two half-lives for the rate of disappearance was assessed. The influence of body weight, body fat and length of treatment upon levonorgestrel clearance was studied through stepwise regression analysis. It was found that the levonorgestrel decay rate after implant removal can be entirely accounted assuming one half-life of 42 +/- 16 h (mean +/- SD; range 13 to 62 h). Stepwise regression analysis showed that levonorgestrel half-life is positively correlated with body weight and not significantly correlated with the length of treatment or body fat. It is concluded that, after long-term administration of levonorgestrel via subdermal implants, the major part of the steroid is cleared from plasma within 96 h and that only trace amounts are detected in the following days.

Adult↗
PubMed1987

Lack of association between spermicide use and trisomy.

It has been suggested that the maternal use of spermicidal contraceptives increases the frequency of certain congenital anomalies, including trisomy, but this issue is in dispute. This controversy led us to examine whether the use of spermicidal contraceptives is associated with an increased risk of fetal trisomy. A questionnaire concerning contraceptive use was completed by 13,729 women who were undergoing prenatal fetal chromosome studies but were as yet unaware of the results. Most women were at increased risk of having a trisomic fetus because of their advanced age. Of 154 fetuses with trisomy, 98 had trisomy 21. For each woman (case) with an affected fetus, four controls were selected from among women with chromosomally normal fetuses, matched for maternal age and medical center. Cases and controls were compared by matched-sample maximum-likelihood logistic regression, to examine the association between fetal trisomy and four measures of spermicide use: periconceptional use, timing of last use, duration of last use, and total lifetime use. No evidence was found for an association, either when all types of trisomy were combined or when trisomy 21 alone was considered. All point estimates of odds ratios relating spermicidal exposure to trisomy were approximately 1, and an effect greater than a twofold increase was excluded with 95 percent confidence in the combined-trisomy group for all measures of spermicide use.

Down Syndrome↗
PubMedDate not supplied

Integrating health services into an MCH-FP program: lessons from Matlab, Bangladesh.

Since 1977 the International Centre for Diarrhoeal Disease Research, Bangladesh, has conducted a field experiment in family planning and MCH in its Matlab research station. The project began with an emphasis on family planning and MCH services were added in stages. This paper uses time series regression methods to address the question of whether the addition of health services contributed to family planning efficacy in Matlab in a program launched with minimal MCH services. The results show that some MCH interventions increased contraceptive prevalence, some decreased it, and others had no effect. The broader significance of these findings for implementing integrated programs is discussed.

Bangladesh↗
PubMed1985

Impact of environmental sanitation and crowding on infant mortality in rural Bangladesh.

In two villages of Bangladesh, 2471 infants born in 1976 and 1977 were followed up for a year to study the impact of environmental sanitation and crowding on their mortality. Neonatal and postneonatal mortality rates in the study cohort were 100 and 75 per 1000 live-births, respectively. Multiple logistic regression analysis was done to estimate the effects of the selected risk factors while controlling for some socioeconomic, demographic, and biological characteristics. Risk of postneonatal mortality (PNNM) in the households which did not use latrines was 3 X 12 times (p less than 0.01) higher than in those which did and 1.5 times (p less than 0.05) higher in the households with 10 or more persons than in smaller households. PNNM in the households which did not use tube-wells (hand-pump) water was higher, but not significantly so, than in those which used tube-well water for all purposes. Neonatal mortality was completely unrelated to the environmental factors investigated.

Bangladesh↗
PubMed1986

Widowhood, sex, labor force participation, and the use of physician services by elderly adults.

This paper investigated the relationships of widowhood, sex, and labor force participation with the use of ambulatory physician services by elderly adults. Data on 18,441 individuals aged 55 and over were taken from the 1978 Health Interview Survey. Hierarchical regression results indicated that although these three factors are related to physician utilization at the zero- and first-partial levels, only sex remained significant when their two- and three-way interactions and other variables from the behavioral model (including living arrangements) were introduced. This suggests that the effects of widowhood and labor force participation are spurious. Widows are simply more likely to live alone and are less likely to work than widowers; those who live alone and do not work are more likely to use health services (and more of them) than those who live with others and are gainfully employed.

Aged↗
PubMed1985

An epidemiological assessment of immunization programme participation in the Philippines.

Because a large proportion of preschool children failed to present for free diphtheria-pertussis-tetanus (DPT) immunizations in a poor, rural area of the Philippines, we undertook an epidemiological analysis of their characteristics. The parents of 159 children were interviewed to determine the demographic, attitudinal, knowledge, and administrative correlates of immunization status. Logistic regression was used to model immunization status. Children were less likely to be immunized if they had a high score on an Adversity Index (composed of measures of the weather, the number of visits the team made, the distance, the appropriateness of the time of day, and miscellaneous problems), if they received health care from a native mother and child health specialist, if a parent was not on the town council, and if pain was an important deterrent. By contrast, many demographic and attitudinal measures that have traditionally been thought to predict health behaviour were not useful discriminators. Recommendations are made for immunization programme management. The general use of this method for programme planning is elaborated.

Adult↗
PubMed1989

Risk factors for low birth weight to adolescent mothers.

The medical records of mothers less than age 20 years who delivered infants weighing 2500 g or less (n = 112) and 2500 g or more (n = 465) were reviewed to identify factors associated with low birth weight. Of the 26 medical and sociodemographic variables examined, four entered a logistic regression model to differentiate low from normal birth weight mothers: five or fewer prenatal visits, a history of a prior low birth weight infant, illness during pregnancy, and trauma (surgery or injury) during pregnancy. The relative risks and adjusted odds ratios for low birth weight associated with these variables ranged from 1.99 to 4.31. Five or fewer visits accounted for the largest proportion of low birth weight, with an etiologic fraction of 0.43. The ability of the model to discriminate between low and normal birth weight mothers was modest; its sensitivity/specificity was 0.73/0.57 in the original sample and 0.62/0.49 in a separate validation sample (n = 329). The model clarifies the factors associated with adolescent low birth weight and suggests that intervention efforts should focus on early pregnancy identification and regular prenatal care.

Adolescent↗
PubMed1995

Use and misuse of oral contraceptives: risk indicators for poor pill taking and discontinuation.

The contraceptive efficacy of oral contraceptives (OCs) depends on their proper and continued use, particularly with lower estrogen preparations. However, few studies have examined why women miss pills or discontinue OCs, and those that do tend to be small and to focus on adolescents. To address the issues of poor OC compliance and early OC discontinuation, we analyzed OC use in a convenience sample of 6,676 women between the ages of 16 and 30 from Denmark, France, Italy, Portugal, and the United Kingdom. Logistic regression was used to examine the independent effect of each factor. Poor compliance was associated with a lack of established routine for pill-taking (relative risk [RR] = 3.3), failure to read and understand written materials that came with the OC package (RR = 2.2), not receiving adequate information or help about OCs from their health care provider (RR = 1.5), and occurrence of certain side effects, including hirsutism (RR = 2.1), nausea (RR = 1.4), bleeding irregularities (RR = 1.3), and breast tenderness (RR = 1.2). Women who were inconsistent OC users, missing one or more pills per cycle, were almost three times as likely to experience an unintended pregnancy while using OCs than were women who took their OCs consistently. Factors that predicted early discontinuation (women who wished to continue contraceptive protection but discontinued OC use) were primarily side effects, including nausea (RR = 2.1), bleeding (RR = 1.9), breast tenderness (RR = 1.8), mood changes (RR = 1.8), and weight gain (RR = 1.4).(ABSTRACT TRUNCATED AT 250 WORDS)

Adolescent↗
PubMed1993

Urban epizootic of rabies in Mexico: epidemiology and impact of animal bite injuries.

From 1 July 1987 to 31 December 1988, a total of 317 animals (91% of which were dogs) were confirmed to have rabies in Hermosillo, Mexico. The median age of rabid dogs was 1 year, 69% were male, and 98% were owned. The epizootic started in the southern areas of the city, rapidly involved the entire city, and persisted mainly in lower socioeconomic status areas. The area of the city and mean household size were significant predictor variables for the population density of rabid dogs around household clusters (Poisson linear regression, P < 0.001 and P = 0.03, resp). Approximately 2.5% of city residents were bitten by dogs in 1987, with the rate of reported dog bite injuries being positively correlated with mean household size and the proportion of households that owned dogs. Visits to the city health centre for evaluation of possible exposures to rabies increased by 135% after the start of the epizootic; approximately 273 per 100,000 city residents were administered a full or partial course of rabies post-exposure prophylaxis in 1987. Children were at greatest risk for exposures to rabies, accounting for 60% of all reported animal bite injuries evaluated at the health centre. Also they were more likely than older persons to have received bite injuries to the head, face, and neck (odds ratio = 21.6, 95% confidence interval = 5.4, 186.5).

Adult↗
PubMed1985

A comparison of direct adjustment and regression adjustment of epidemiologic measures.

Although regression adjustment can provide a useful alternative to direct adjustment, especially when data are sparse, many researchers are unaware that adjusted summary measures can be easily derived from regression coefficients. In a non-technical discussion with examples, the direct adjustment procedure is compared with three methods of regression adjustment based on analysis of covariance models: the conditional prediction method, the stratified prediction method, and the marginal prediction method. Both the stratified prediction and direct adjustment methods yield summary measures that are weighted averages of stratum-specific measures, while adjusted measures from the conditional prediction method are similar to stratum-specific estimates. In contrast to the other adjustment procedures, which can use internal or external weights, the marginal prediction method always gives an internally adjusted measure. Under certain conditions, the three regression adjustment procedures produce identical results. Major advantages of direct adjustment include computational simplicity and relatively few statistical assumptions. Regression adjustment, however, is more convenient for statistical tests for interactions and group differences, and often precludes the need to categorize continuous variables, so that problems with empty strata are avoided.

Age Factors↗
PubMed2004

Complete pathologic response following preoperative chemoradiation therapy for middle to lower rectal cancer is not a prognostic factor for a better outcome.

PURPOSE: The aim of this study was to evaluate factors associated with pathologic tumor response following pre-operative chemoradiation therapy, and the prognostic impact of pathologic response on overall and disease-free survival. METHODS: Between 1994 and 2002, 132 patients underwent chemoradiation therapy followed by surgery for middle to lower rectal cancer. After excluding 26 cases (metastatic cancer, n = 13; nonradical surgery, n = 6; local excision procedure, n = 4; non-5-fluorouracil-based chemotherapy, n = 2; incomplete data on preoperative chemoradiation therapy regimen used, n = 1), the remaining 106 patients were included in the study. Variables considered were the following: age, gender, tumor location, pretreatment T and N stage, modality of 5-fluorouracil administration, total radiotherapy dose delivered, chemoradiation therapy regimen used (Regimen A: chemotherapy (bolus of 5-fluorouracil and leucovorin, days 1-5 and 29-33) + radiotherapy (45 Gy/25 F/1.8 Gy/F); Regimen B: chemotherapy (5-fluorouracil continuous venous infusion +/- weekly bolus of carboplatin or oxaliplatin) + radiotherapy (50.4 Gy/28 F/1.8 Gy/F)), time interval between completion of chemoradiation therapy and surgery, postoperative chemotherapy administration, surgical procedures, pT, pN, and pTNM stage, and response to chemoradiation therapy defined as tumor regression grade, scored from 1 (no tumor on surgical specimen) to 5 (absence of regressive changes). Statistical analysis was performed by means of logistic regression analysis (Cox's model for overall and disease-free survival). RESULTS: Median age of the 106 patients was 60 (range, 31-79) years and the male:female ratio, 66:40. Median distance of tumor from the anal verge was 6 (range, 1-11) cm. Pretreatment TNM stage, available in 104 patients, was cT3T4N0, n = 41; cT2N1, n = 9; cT3N1, n = 39; and cT4N1, n = 17. The median radiotherapy dose delivered was 50.4 (range, 40-56) Gy; 58 patients received 5-fluorouracil by continuous venous infusion, and carboplatin with oxaliplatin was added to the chemotherapy schedule in 71 cases. Patients were given Regimen A in 47 cases and Regimen B in 59. The median interval between chemoradiation therapy and surgery was 42.5 (range, 19-136) days, and 94 patients underwent a sphincter-saving procedure. Tumor regression grade, available in 104 cases, was 1, n = 19; 2, n = 18; 3, n = 15; 4, n = 13; and 5, n = 39. At a median follow-up of 42 (range, 1-110) months, 11 patients had died, and 95 were alive. None of the patients had local recurrences, but 13 had distant recurrences. At logistic regression analysis, the chemoradiation therapy regimen used was the only independent predictor of tumor response following preoperative chemoradiation therapy (odds ratio = 0.29, 95% confidence interval = 0.13-0.67, P = 0.003). At Cox's regression analysis, pretreatment T stage was the only independent prognostic factor for both disease-free survival (relative risk = 7.13, 95% confidence interval = 2.3-21.8, P = 0.001) and overall survival (relative risk = 4.83, 95% confidence interval = 1.1-19.9, P = 0.029). CONCLUSIONS: Tumor response following preoperative chemoradiation therapy is mainly related to the preoperative regimen used. For patients receiving preoperative chemoradiation therapy, pretreatment T stage, but not tumor response to preoperative chemoradiation therapy, is prognostic for outcome (both disease-free and overall survival).

Adenocarcinoma↗
PubMed1993

A comparison of empirical models on determinants of infant mortality: a cross-national study on Africa.

The goal of achieving health for all by the year 2000 has instigated numerous studies on the determinants of health. In this paper, we re-evaluate two models in which infant mortality--across twenty-eight low- and middle-income African states--is explained by socioeconomic, demographic, medical, environmental and political factors. The results indicate that the GNP per capita, school population as a percentage of the population under 19, population density and the percentage of the population with access to health care together explained 80% of the variations in infant mortality in the sample study. Apart from the GNP per capita and the school population as a percentage of the population under 19 which were negative and statistically significant, variables of importance for health policy, e.g. female literacy rate, water supply, food aid, calorie supply, health care expenditure and the degree of urbanization carried a negative sign but were nonsignificant. We interpreted the above thus: a reduction in infant mortality is feasible only with changes on diverse fronts rather than by marginal improvements in a few determining factors. A comparative test of the replicated model and our proposed model has shown that our model produced a better theoretical and statistical fitting than did the replicated model.

Africa↗
PubMed2002

Multiple stressor effects on benthic biodiversity of Chesapeake Bay: implications for ecological risk assessment.

In natural aquatic systems, there is frequently overlap in the spatial distribution of both natural and anthropogenic stressors, particularly at regional geographic scales. Yet the proportional risk associated with individual stressors, their cumulative effects and the manner in which they interact to affect aquatic ecology is frequently unknown, limiting the robustness of multiple-stressor ecological risk assessments (ERA). The current study used historical environmental monitoring data (1984-1999) to identify a combination of natural and anthropogenic stressors that best accounted for observed patterns of benthic biodiversity in Chesapeake Bay. Geographic information systems (GIS) were used to geographically link spatially heterogeneous databases for benthic biodiversity, water quality and sediment toxicant concentrations. Single and multiple variable regression techniques were subsequently used to develop a statistical model to explain observed patterns of benthic biodiversity. Combinations of natural stressors alone accounted for as much as 34% of the variation in benthic biodiversity, and combinations of anthropogenic toxicants accounted for as much as 48% of the variation. The consideration of both natural and anthropogenic stressors resulted in a statistical model that accounted for approximately 73% of the observed variation in benthic biodiversity of Chesapeake Bay. These results suggest that benthic biodiversity in Chesapeake Bay is a function of complex interactions among water quality characteristics and anthropogenic toxicants. Therefore, new risk assessment methodologies are required to assess the risk of multiple stressors at regional scales.

Animals↗
PubMed1980

Associations between cancer incidence and alcohol/cigarette consumption among five ethnic groups in Hawaii.

The average annual age-adjusted incidence rates of cancer for 15 sites were determined for 10 ethnic-sex groups in Hawaii. Consumption rates for cigarettes, beer, wine and hard liquor were also determined for the same 10 groups based on personal interview of a sample of 9920 individuals. Covariance analysis was used to adjust each exposure variable for the other three, and the cancer incidence rates were then linearly regressed on these covariance-adjusted consumption rates. Statistically significant regression coefficients were found for cancer of the tongue/mouth, pharynx, larynx, pancreas, lung, kidney and bladder regressed on cigarette consumption. Eight cancer sites, including tongue/mouth, pharynx, larynx, oesophagus, stomach, pancreas, lung and kidney, had significant positive regression coefficients for beer consumption which could not be explained by outlying values on the scattergram. Significant associations were also suggested between wine consumption and pharyngeal cancer and between hard-liquor consumption and pharyngeal, laryngeal and possibly brain cancer. No association was found between beer consumption and colorectal cancer. Multiple regression analysis with sex, cigarettes and alcoholic beverage as independent variables consistently found sex to be least important in determining cancer risk. This study supports the hypothesis that beer consumption may play a role in cancer risk for several sites. It is suggested that future studies of alcoholic beverages and cancer should examine not only types of alcoholic beverages, but individual brands of each type in an attempt to identify cancer risk due to carcinogens in only certain brands.

Alcohol Drinking↗
PubMed2005

Variable selection for marginal longitudinal generalized linear models.

Variable selection is an essential part of any statistical analysis and yet has been somewhat neglected in the context of longitudinal data analysis. In this article, we propose a generalized version of Mallows's C(p) (GC(p)) suitable for use with both parametric and nonparametric models. GC(p) provides an estimate of a measure of model's adequacy for prediction. We examine its performance with popular marginal longitudinal models (fitted using GEE) and contrast results with what is typically done in practice: variable selection based on Wald-type or score-type tests. An application to real data further demonstrates the merits of our approach while at the same time emphasizing some important robust features inherent to GC(p).

Bias↗
PubMed1991

Variation in torus mandibularis prevalence in Norway. A statistical analysis using logistic regression.

The prevalence of torus mandibularis was assessed in two groups of dental patients, altogether 2010 individuals over 10 yr of age: 1181 individuals native to the Lofoten Islands in North Norway, situated at 68 degrees latitude; and 829 patients indigenous to Gudbrandsdal, an inland district in the southeastern part of the country at 61 degrees latitude. Both groups were supposed to be of the same Caucasian stock and, therefore, to have similar genetic predisposition to torus on the average. The following observations were found: 1) the prevalence of torus mandibularis was much greater in Gudbrandsdal than in Lofoten (P much less than 0.001); 2) the prevalence decreased among persons above 50 yr of age as compared with those of the age classes 10-49 yr (P less than 0.01); and 3) it was smaller among women than men (P less than 0.05), mainly due to such a decrease in Lofoten. In a recent investigation of people living in Gudbrandsdal the fraction of the variation of torus that was attributable to genetic differences was estimated as about 30%, whereas approximately 70% of the causes seemed to be ascribable to environmental influences in terms of occlusal stress. It is suggested that dietary habits and number of existing teeth seemed to be environmental variables with an influence on the observed variation of torus prevalence between geographical regions, age classes, and sexes. The question about a possible sexual difference as to the genetic component of liability to torus mandibularis was outside the scope of the present study.

Adolescent↗