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 703 records · Page 39Linked to original sources
PubMedDate not supplied

Hepatic and gastrointestinal effects in an occupational cohort exposed to 2,3,7,8-tetrachlorodibenzo-para-dioxin.

OBJECTIVE: To examine the effect of occupational exposure to substances contaminated with 2,3,7,8-tetrachlorodibenzo-para-dioxin (TCDD) on the liver and gastrointestinal system. DESIGN: A medical survey. PARTICIPANTS: The exposed participants were employed at two chemical plants more than 15 years earlier in the manufacture of sodium trichlorophenol and its derivatives. The reference group consisted of individuals with no occupational exposure to phenoxy herbicides and who lived within the communities of the workers. A total of 281 workers and 260 unexposed referents participated in the medical study. MEASUREMENTS AND MAIN RESULTS: The workers had substantial exposure to substances contaminated with TCDD, as evidenced by a mean serum TCDD level, lipid adjusted, of 220 pg per gram of lipid compared with a mean of 7 pg per gram of lipid in the referents. Compared with the unexposed reference group, workers had a statistically significantly elevated risk for an out-of-range gamma-glutamyltransferase (GGT) level (odds ratio, 2.27; 95% confidence interval, 1.17 to 4.39 [unadjusted for confounders]). In multivariate analyses run with logistic regression, a statistically significant interaction was found between TCDD exposure and lifetime alcohol consumption, indicating that the elevated risk for an out-of-range GGT was confined to those workers with a history of alcohol consumption and that the risk among the alcohol-consuming workers for an out-of-range GGT increased with increasing TCDD level. No difference was found between workers and referents for any of the other liver and gastrointestinal outcomes of interest. CONCLUSIONS: This study found no evidence of an elevated risk for clinical hepatic or gastrointestinal disease in a group of workers with high exposure to TCDD. However, TCDD-exposed workers with a history of sufficient alcohol consumption were found to have a statistically significantly elevated risk for an out-of-range GGT compared with referents.

Cross-Sectional Studies↗
PubMed2005

Domestic violence and women's employment: fixed effects models of three waves of women's employment study data.

Domestic violence can interfere with women's ability to work, and may result in loss of welfare benefits and poorer economic outcomes. Previous studies showing no effect of domestic violence on employment could be a result of the failure to control for some individual characteristics; therefore we use fixed-effects models with three waves of Women's Employment Study (WES) data to control for unobservable time-invariant individual-specific characteristics. Included in our analyses were 598 women, from an urban county in Michigan who were on the welfare rolls in February of 1997, all of whom completed 3 waves of interviews. Our fixed effects regression results show that domestic violence significantly reduced the annual work hours of a respondent. Mental and physical health problems do not completely mediate this relationship. Our results support efforts to address domestic violence within the welfare system.

Adult↗
PubMed2004

Randomized trial of intraportal and/or systemic adjuvant chemotherapy in patients with colon carcinoma.

BACKGROUND: 5-fluorouracil-based adjuvant chemotherapy after surgical resection of colon cancer is standard treatment. However, the choice of best delivery route--that is, systemic (i.e., intravenous or oral) or regional (i.e., intraportal, intraperitoneal, or hepatic arterial infusion)--has been controversial. In a randomized clinical trial of patients with colon cancer, we compared the benefits of chemotherapy delivered by these routes individually or in combination. METHODS: From April 2, 1992, through April 30, 1998, 1084 eligible patients with Dukes' stage B or C colon carcinoma were randomly assigned: 369 patients to the IP regimen (continuous portal vein infusion of 5-fluorouracil at 500 mg/m2 of body surface daily and heparin at 5000 IU daily for 7 consecutive days, beginning on the day of surgery), 358 patients to the SY regimen (six 28-day courses of systemic leucovorin at 100 mg/m2 daily on days 1 through 5 followed by systemic bolus 5-fluorouracil at 370 mg/m2 daily on days 1 through 5, with treatment initiated 15-35 days after surgery), and 357 patients to the IP+SY regimen (the IP regimen followed by the SY regimen, with the same scheduling). Primary survival was analyzed with the log-rank statistic and a Cox multivariable regression model. All statistical tests were two sided. RESULTS: At a median follow-up time of 99 months, 389 events (recurrences, second malignancies, or deaths) had occurred, and 361 patients died. Sites of first recurrences were similar among the three arms. At 5 years, overall and event-free survival rates were similar among those on the IP (74% and 68%, respectively), SY (78% and 71%), and IP+SY (73% and 67%) regimens. When compared with the group on the SY regimen, the risk for death associated with the IP regimen (hazard ratio [HR] = 1.05, 95% confidence interval [CI] = 0.82 to 1.36) was similar to that associated with the IP+SY regimen (HR = 1.12, 95% CI = 0.78 to 1.45) (P =.69), as were the risks for first event (HR = 1.07, 95% CI = 0.84 to 1.37 and HR = 1.10, 95% CI = 0.86 to 1.41, respectively) (P=.74). CONCLUSION: Overall and event-free survival rates were similar in all three arms. The combined regimen was no better than either single regimen alone.

Adenocarcinoma↗
PubMed2005

Systematic aortic and pelvic lymphadenectomy versus resection of bulky nodes only in optimally debulked advanced ovarian cancer: a randomized clinical trial.

BACKGROUND: The role of systematic aortic and pelvic lymphadenectomy in patients with optimally debulked advanced ovarian cancer is unclear and has not been addressed by randomized studies. We conducted a randomized clinical trial to determine whether systematic aortic and pelvic lymphadenectomy improves progression-free and overall survival compared with resection of bulky nodes only. METHODS: From January 1991 through May 2003, 427 eligible patients with International Federation of Gynecology and Obstetrics (FIGO) stage IIIB-C and IV epithelial ovarian carcinoma were randomly assigned to undergo systematic pelvic and para-aortic lymphadenectomy (n = 216) or resection of bulky nodes only (n = 211). Progression-free survival and overall survival were analyzed using a log-rank statistic and a Cox multivariable regression analysis. All statistical tests were two-sided. RESULTS: After a median follow-up of 68.4 months, 292 events (i.e., recurrences or deaths) were observed, and 202 patients had died. Sites of first recurrences were similar in both arms. The adjusted risk for first event was statistically significantly lower in the systematic lymphadenectomy arm (hazard ratio [HR] = .75, 95% confidence interval [CI] = 0.59 to 0.94; P = .01) than in the no-lymphadenectomy arm, corresponding to 5-year progression-free survival rates of 31.2 and 21.6% in the systematic lymphadenectomy and control arms, respectively (difference = 9.6%, 95% CI = 1.5% to 21.6%), and to median progression-free survival of 29.4 and 22.4 months, respectively (difference = 7 months, 95% CI = 1.0 to 14.4 months). The risk of death was similar in both arms (HR = 0.97, 95% CI = 0.74 to 1.29; P = .85), corresponding to 5-year overall survival rates of 48.5 and 47%, respectively (difference = 1.5%, 95% CI = -8.4% to 10.6%), and to median overall survival of 58.7 and 56.3 months, respectively (difference = 2.4 months, 95% CI = -11.8 to 21.0 months). Median operating time was longer, and the percentage of patients requiring blood transfusions was higher in the systematic lymphadenectomy arm than in the no-lymphadenectomy arm (300 versus 210 minutes, P<.001, and 72% versus 59%; P = .006, respectively). CONCLUSION: Systematic lymphadenectomy improves progression-free but not overall survival in women with optimally debulked advanced ovarian carcinoma.

Adult↗
PubMed1993

Changes in the determinants of fertility in Korea: analyses of pregnancy intervals and outcomes.

Korea completed the whole process of what is called the demographic transition to a low fertility and mortality level with the successful implementation of the national family planning program in 1962, and this has been accompanied by rapid socioeconomic development. Most of the fertility decline was due to a rising age at marriage and to lower marital fertility. The national family planning program, combined with the widespread practice of induced abortion, has placed an important role in reducing marital fertility, particularly among older women. This paper aims, therefore, to examine the determinants of fertility and their changes over time in an effort to suggest future population policy directions for Korea. The analysis is divided into 2 parts-estimation of pregnancy intervals by applying the proportional hazards model, and estimation of the determinants of fertility by adopting the logistic regression model to find out whether a pregnancy terminates in a live birth or in an abortion. In both analyses, the sex was included as the main explanatory variable. A woman's education has been shown to have a significant effect on delaying the timing or on the wife's age at first pregnancy, but its effect on the pace of subsequent pregnancies is much smaller and often positive. On the other hand, the woman's education has a consistently positive effect on the probability of a pregnancy ending in an abortion although the effect shows a steady decline over time. Form first parity, the sex composition of previous children stands out consistently as the most important factor in deciding both the pace of pregnancy and its outcome. The pregnancy risks of the women with sons are reduced by almost 50% at the second and third parities. The probability of a pregnancy ending in an abortion also increases substantially when parents already have a son. The decline of the desired family size but the sustained strong son preference has made the sex of children a more important factor in the determination of Korean fertility. The woman's education, on the other hand, has become a less important factor.

Abortion, Induced↗
PubMed2006

"Hang Up Your Pocketbook" -- an easy intervention for the granny syndrome: grandparents as a risk factor in unintentional pediatric exposures to pharmaceuticals.

CONTEXT: Although the circumstances are not well studied, grandparents' medications account for 10% to 20% of unintentional pediatric intoxications in the United States. OBJECTIVES: To characterize circumstances leading to and outcomes from pediatric pharmaceutical exposures. To identify preventable risk factors associated with this pattern of injury, referred to as the "granny syndrome." DESIGN, SETTING, AND PARTICIPANTS: Retrospective review of records of telephone calls made to a certified regional poison control center in the United States. Records were analyzed for all calls concerning children aged 6 years or younger who were exposed to grandparents' medication(s). For statistical analysis, regression and chi(2) analysis as well as Fisher exact tests were used. The sample size provided 80% power to detect a 10% difference at the 5% level of significance. Statistical significance was set at P< or =.05. MAIN OUTCOMES MEASURED: Use of child-resistant containers (CRCs), the location of pharmaceuticals prior to pediatric ingestion, and drug classes involved (eg, analgesics, cardiovascular drugs). RESULTS: Of the 200 incidents analyzed, 90 (45%) cases involved CRCs, and 110 (55%) involved containers that were not child resistant. For these incidents, the average age of the child was 18.8 months; the grandparent was aged on average 58.7 years. Most medications had been placed on tables or countertops (91 [46%]), low shelves (57 [29%]), or in pocketbooks (34 [17%]). The type of container in which the pharmacologic agent was stored (CRC vs non-CRC) was not statistically significant (P>.1). Ease of access to medication, regardless of the type of container used, was the only statistically significant outcome (P<.001). In the present study, accidental pediatric exposures most frequently involved cardiovascular (90 [45%]), analgesic (84 [42%]), and psychotropic (32 [16%]) medications. CONCLUSION: Pediatric exposure to pharmaceutical agents is a preventable cause of injury. Physicians have an important opportunity to assist in preventing pediatric pharmaceutical exposures by instructing parents and grandparents on how to better limit children's access to medications as an essential component to enhance child safety.

Accidents, Home↗
PubMed1996

[The action of noradrenaline on cardiac energetics in the rat].

Noradrenaline induced a parallel shift of the energy expenditure regression with no statistically significant change of the regression coefficient in isolated rat heart. It also increased the energy expenditure when acting upon and arrested heart with the energy exchange level approximating a non-contractile energy expenditure in the working heart preparation. This finding seems to be mainly due to changes in the non-contractile energy exchange.

Animals↗
PubMed1989

Cigarettes and alcohol as independent risk factors for colonic adenomas.

Healthy adults completed smoking and alcohol consumption questionnaires before colonoscopies, which were performed because of occult blood in the stool or prior barium enema suggesting polyps. Subjects with adenomas at colonoscopy (n = 102) were compared with colonoscopy-negative controls (n = 89). In univariate analyses, age (p less than 0.05), male sex (p less than 0.005), cumulative smoking (p less than 0.0001), and cumulative beer consumption (p less than 0.005) were associated with adenomas. The association of smoking with adenomas was stronger in younger subjects. The association of beer with adenomas was stronger in older subjects. Logistic regression confirmed statistically significant associations of smoking (odds ratio for greater than 40 pack-years = 3.31; confidence intervals 1.41, 7.81) and beer consumption (odds ratio for greater than 40 beer-years = 2.64; confidence intervals 1.10, 6.32) with adenomas. These results suggest that smoking and beer consumption are independent risk factors for colonic adenomas.

Adenoma↗
PubMed2004

Measurement of rheology of distiller's grain slurries using a helical impeller viscometer.

Current research is focused on developing a process to convert the cellulose and hemicellulose in distiller's grains into fermentable sugars, increasing both ethanol yield and the amount of protein in the remaining solid product. The rheologic properties of distiller's grain slurries were determined for concentrations of 21, 23, and 25%. Distiller's grain slurries are non-Newtonian, heterogeneous fluids subject to particle settling. Traditional methods of viscosity measurement, such as cone-and-plate and concentric cylinder viscometers, are not adequate for these fluids. A helical impeller viscometer was employed to measure impeller torque over a range of rotational speeds. Newtonian and non-Newtonian calibration fluids were utilized to obtain constants that relate shear stresses and shear rates to the experimental data. The Newtonian impeller constant, c, was 151; the non-Newtonian shear rate constant, k, was 10.30. Regression analysis of experimental data was utilized for comparison to power law, Herschel-Bulkley, and Casson viscosity models with regression coefficients exceeding 0.99 in all cases.

Biotechnology↗
PubMed2003

Exploring the relationship between surrogates and clinical outcomes: analysis of individual patient data vs. meta-regression on group-level summary statistics.

There has been an increasing interest in exploring the relationship between a surrogate and a clinical outcome. Two different statistical approaches have been taken by researchers to quantify the treatment effect on the clinical outcome explained by the surrogate endpoint: 1) analysis based on individual patient data (IPD), and 2) meta-regression based on summary statistics from published literature. An analysis based on IPD models the associations between the surrogate and clinical outcome for patients directly and is able to adjust for patient-level covariates. A meta-regression models the trial-level associations using group-level summary statistics and trial-level covariates. The results from these two approaches can be quite disparate and researchers may reach different conclusions on scientific questions that they wish to answer. We demonstrate that the typical summary statistics, such as group means and event counts, do not provide a set of sufficient statistics for estimating the underlying relationship between the surrogate and clinical outcome for patients. Consequently, the associations derived from meta-regression do not necessarily reflect the causal relationship for patients and should be interpreted with caution. A meta-analysis of antiresorptive agents for osteoporosis serves to illustrate the magnitude of differences between the two approaches.

Humans↗
PubMed1992

Robust bivariate errors-in-variables regression and outlier detection.

A bivariate regression model is introduced where both variables are subject to error. The structural regression line is equivariant against interchanging coordinates and permits bivariate calibration, i.e. the prediction of one variable by means of the other. Maximum likelihood and robust parameter are estimated, based on order statistics. Residual analysis and outlier detection are performed. The model is applied to the comparison of clinical chemical analytical methods.

Hematocrit↗
PubMed2002

Background beliefs in Bayesian inference.

We report five experiments in which the role of background beliefs in social judgments of posterior probability was investigated. From a Bayesian perspective, people should combine prior probabilities (or base rates) and diagnostic evidence with equal weighting, although previous research shows that base rates are often underweighted. These experiments were designed so that either piece of information was supplied either by personal beliefs or by presented statistics, and regression analyses were performed on individual participants to assess the relative influence of information. We found that both prior probabilities and diagnostic information significantly influenced judgments, whether supplied by beliefs or by statistical information, but that belief-based information tended to dominate the judgments made.

Bayes Theorem↗
PubMed2006

Early endoscopic treatment of acute inflammatory airway lesions improves the outcome of postintubation airway stenosis.

OBJECTIVES/HYPOTHESIS: The objective of this study was to compare the results of treating acute postintubation stenotic airway lesions with the results of treating mature lesions and to assess whether early intervention improves the outcome. METHODS: Patients without previous surgery, treated for postintubation airway stenosis, were reviewed. Those with airway lesions presenting within weeks of intubation (n = 11) were treated with intralesional steroids, laser reduction, and balloon dilatation. Patients with mature airway lesions (n = 20) were initially treated with laser, balloon dilatation, and topical mitomycin C. Preoperative patient and lesion characteristics, details of the treatment, and intervention-free intervals were recorded. Data were compared with unpaired Student t test or chi test as appropriate, and intervention-free intervals were compared with log-rank statistics. Cox regression was used to identify independent predictors of intervention-free interval. RESULTS: There were 18 males and 13 females. The average age at presentation was 47 +/- 2.7 years. The two groups were statistically comparable for demographic and lesion characteristics. Patients treated for acute injury required significantly fewer interventions (P < .03), the majority being treated with a single treatment. They had a significantly longer intervention-free interval (P < .02; log-rank analysis) and did not require external laryngotracheal reconstruction (P < .001) compared with patients treated for mature fibrotic scars leading to airway stenosis. CONCLUSIONS: Early treatment of acute fibroinflammatory airway lesions has the potential to favorably modify the natural history of postintubation tracheal injury, raising the tantalizing possibility of an early cure. These findings have significant implications for early identification, referral, and treatment of postintubation tracheal stenosis and need to be confirmed with further studies.

Acute Disease↗
PubMed1996

The effect of managed care on ICU length of stay: implications for medicare.

OBJECTIVE: To determine whether insurance status (managed care vs traditional commercial and Medicare) influences resource consumption (as measured by length of stay [LOS]) in the intensive care unit (ICU). DESIGN: Retrospective analysis of the 1992 Massachusetts state hospital discharge database, using prospectively developed and validated risk-stratification models. SETTING: All nonfederal hospitals in Massachusetts. SUBJECTS: Of all adult hospitalizations where an ICU stay was incurred (n=104270), we selected those covered by 1 of 4 payer groups (n=88050): (1) commercial fee-for-service (patients aged <65 years); (2) commercial managed care (patients aged <65 years); (3) traditional Medicare (patients aged >/=65 years); and (4) Medicare-sponsored managed care (patients aged >/=65 years). MAIN OUTCOME MEASURE: Mean ICU LOS. ANALYSIS: The ICU LOS regression models were constructed using split-halves validation to adjust for differences in age, sex, severity of illness, diagnosis, discharge status, and payer. Separate models were constructed for those younger than 65 years and those aged 65 years or older. Robustness of the models was explored using goodness of fit and correlation. The effect of payer on hospital mortality was also explored using logistic regression. Observed minus predicted mean ICU LOS and mortality rates were correlated with managed care penetration at the hospital level. RESULTS: The ICU LOS models performed well (R2=0.84 and R2L [likelihood ratio statistic]=0.92 for the development set, and R2=0.83 and R2L=0.89 for the validation set). Significant covariables affecting LOS included age, severity of principal illness, comorbidity, reason for admission, and discharge status (P<.001 for each). Among the cohort younger than 65 years (n=27805), although unadjusted mean ICU LOS was shorter (2.9 vs 3.43 days; P<.05) for those covered by managed care organizations, payer status had no independent effect on ICU LOS (P=.48). Among those older than 65 years, there was neither a difference in unadjusted ICU LOS (3.94 vs 3.88 days; P>/=.05) nor an independent effect of payer on ICU LOS (P=.35). Unadjusted mortality was lower among managed care patients (3.9% vs 5.1% in patients aged <65 years [P<.05] and 8.7% vs 12.1% in patients aged > or = 65 years [P<.05]). Age, severity of principal diagnosis, comorbidity, and reason for admission significantly influenced mortality (P<.001). After controlling for these factors with the mortality model (R2L=0.92 and 0.89, C statistic [12 df]=8.45 and 17.58, and P=.75 and .13 [where a large P reflects good agreement] for the development and validation sets, respectively), payer continued to have a small but significant effect on mortality (odds ratios ranging from 1.67 at 0.1% probability of death to 1.11 at 30% probability of death.) Managed care penetration among the commercially insured varied across hospitals (n=82) from 0% to 68%. There was no correlation between managed care penetration and either ICU LOS (R2=0.04; P=.09) or mortality (R2=0.0; P=.88). CONCLUSIONS: Though patients covered under managed care consume fewer ICU resources, this appears to be primarily attributable to a difference in patient-related factors. Thus, as managed care case mix changes in the future to include sicker and older patients, the initial advantages of reduced resource consumption may diminish.

Adult↗