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 1,153 records · Page 64Linked to original sources
PubMed1995

Urogenital infection in preeclampsia.

OBJECTIVE: The purpose of this study was to determine whether the incidence of urinary tract infections and postpartum endometritis were increased in preeclamptic pregnancies. METHOD: We conducted a retrospective study of 13852 pregnant women, using a perinatal database at The Johns Hopkins Hospital, over the past 5 years. The incidence of urinary tract infections and postpartum endometritis was analyzed using the chi-squared test and logistic regression analysis. Statistical significance was set at P < 0.05. RESULTS: There were 345 (2.5%) mild preeclamptics and 440 (3.2%) severe preeclamptics. The incidence of urinary tract infections and postpartum endometritis in preeclamptic patients was significantly higher than that in non-hypertensive pregnant patients. After controlling for confounding variables, severe preeclampsia was still found to be an independent significant risk factor for both urinary tract infections and postpartum endometritis. CONCLUSION: Our data show a significant increase in urogenital infection in preeclamptic pregnancy. This may reflect higher rates of underlying renal disease and placental bed abnormalities occurring in preeclampsia.

Adult↗
PubMed1983

Effect of 131I on the anemia of hyperthyroidism.

Data from the National Thyrotoxicosis Therapy Follow-Up Study (NTTFS) are presented here to document the existence of anemia in hyperthyroidism, a mild and reversible anemia that is simultaneously ameliorated with reversal of the hyperthyroid state. Among 20,600 women entered into the NTTF study with no previous history of hematological disorders, the prevalence of anemia was found to range from 10-15%, appearing to be higher in those selected for treatment with 131I when compared to those selected for surgery. An attempt is made to verify the recent hypothesis that thyroid hormone levels in the supraphysiologic range may suppress erythrogenesis. Two statistically significant regression models are consistent with a hypothesis of thyrotoxic bone marrow suppression. However, both associations are weak enough to suggest that some other physiologic improvement underlies the amelioration of anemia when hyperthyroidism is reversed. The degree of improvement in hematological status is similar for women in both treatment groups. Among 4464 women for whom serial hematological tests are obtained, over 3/4 of anemic patients are no longer anemic after an average 6.2 yr of follow-up. Clinicians are reassured that radioactive iodine exposure causes no further insult to the bone marrow, no matter what the cumulative dosage. The highly fractionated low dose bone marrow exposures to radiation account for the minimal hematological risks of 131I treatment.

Anemia↗
PubMed1992

Modelling the use of health services by populations of small areas to inform the allocation of central resources to larger regions.

Modelling the use of health services by local populations can inform the allocation of central resources to larger regions. At aggregate, e.g. regional, level, use is largely driven by supply and such analysis can be misleading. At small area level, however, populations effectively compete for services and their different relative use of services, after allowing for varying availability, reflects variations in their relative need. Empirical quantitative estimates can thus be made of the net relation of factors such as mortality or social deprivation to relative need for health services. These estimates can then be used in weighted capitation formulae for allocating resources to larger regions.

Budgets↗
PubMed1987

Religious observance and plasma lipids and lipoproteins among 17-year-old Jewish residents of Jerusalem.

The association of religious observance and plasma lipids and lipoproteins was studied in a sample of 673 Jewish residents of Jerusalem ages 17-18 years. Religious observance was classified according to the parents' ranking of their perceived degree of religiosity. The study group included children whose parents were orthodox Jews who pedantically observed religious commandments, children of traditional parents who observed some of the rules, and children of nonobservant secular parents. Plasma levels of cholesterol, triglyceride, and low-density lipoprotein were higher in secular children than in the orthodox group. These associations were independent of sex, origin, social class, body mass, and season. High-density lipoprotein cholesterol was somewhat higher in the orthodox group than in the secular children although this difference was not statistically significant. Regression analysis showed that offspring's environment and parental phenotype were the most important predictors of lipid concentrations in adolescents. The association of religious observance with plasma lipids and lipoproteins, however, was independent of parental phenotype lipid values and the contribution of offspring and parents' environment. These findings are consistent with similar differences in plasma lipids described previously among the parents, as well as the lower incidence of myocardial infarction in the orthodox religious group, which has been shown in the Israeli adult population.

Adolescent↗
PubMed1992

Hypernatraemia surveillance during a national diarrhoeal diseases control project in Egypt.

The nationwide introduction of oral rehydration therapy to Egypt has led to improvement in diarrhoea case management and a fall in infant and child mortality. With the wider use of oral rehydration solution (ORS) prepared from packets, the incidence of hypernatraemia (serum sodium greater than 150 mmol/l) in inpatients with dehydration seen at Abu El-Reeche Hospital, Cairo, increased between 1980 and 1984. Systematic surveillance of hypernatraemia in the outpatient rehydration unit began in late 1984, and we report trends in hypernatraemia and analyses of key variables affecting its incidence in dehydrated children. In 1980, 17 of 100 children sampled had hypernatraemia and 2 had severe hypernatraemia (ie, serum sodium greater than 165 mmol/l). The frequency in inpatients peaked at 49% of 222 children in 1984 (19% with severe hypernatraemia). Between 1986 and 1989, at least 1000 dehydrated outpatients were surveyed each year; by 1989 the incidence of hypernatraemia had fallen to around 10% (2% severe hypernatraemia). The rise and decline coincided with increasing use of ORS and then increasing ability of mothers to mix the solution correctly. Hypernatraemia was positively related to the quantity of ORS taken, severity of dehydration, nutritional status, and the cooler season, and negatively related to age and duration of diarrhoea. Explanations for our findings include improved use of ORS and better case-management. Good practice promoted through the mass media has facilitated these changes; if the standard of ORS use is not maintained, there may be a case for reducing the sodium content of ORS.

Child, Preschool↗
PubMed1990

The dynamic relationship between low birthweight and induced abortion in New York City. An aggregate time-series analysis.

We use a vector autoregression to examine the dynamic relationship between the race-specific percentage of pregnancies terminated by induced abortion and the race-specific percentage of low-birthweight births in New York City. With monthly data beginning in 1972, we find that induced abortion explains low birthweight for blacks, but not for whites. There is no evidence of feedback from low birthweight to induced abortion. The findings suggest that unanticipated decreases in the percentage of pregnancies terminated by induced abortion would worsen birth outcomes among blacks in New York City.

Abortion, Induced↗
PubMed1991

On the measurement of horizontal inequity in the delivery of health care.

This paper offers a critical appraisal of the various methods used to date to investigate inequity in the delivery of health care. It concludes that none of the methods used to date is particularly well equipped to provide unbiassed estimates of the extent of inequity. It also concludes that Le Grand's (1978) approach is likely to point towards inequity favouring the rich even when none exists. The paper offers an alternative approach, which builds on the approaches to date but seeks to overcome their deficiencies.

Acute Disease↗
PubMed1994

A new test for supplier-inducement and application to the Canadian market for dental care.

The hypothesis that dentists do not induce demand for their services is tested using reduced form estimates of the price elasticity of demand. If demand is autonomous, shifts in supply for whatever reason should generate equivalent estimates provided access costs change proportionately with shifts in supply. If demand inducement is present demand can appear to be very elastic, or very inelastic, depending on what is causing the shift in supply. Each of three applications of this test, conducted in the context of jointly estimated fee and quantity equations using annual Canadian data for 1956-1989, rejects the no inducement hypothesis.

Canada↗
PubMed1995

Is moderate alcohol use related to wages? Evidence from four worksites.

Moderate alcohol users (approximately two drinks per day on average) have recently been shown to have a lower risk of coronary heart disease relative to abstainers and heavy drinkers. Conversely, a few studies have found that alcoholism is associated with greater unemployment and lower earnings. But, little research has examined the differential effects of drinking levels on job compensation. We utilize a newly created database on employees at four worksites to test for a nonlinear relationship between alcohol use and wages. We also examine whether alcohol use affects wages indirectly through educational achievement, marital status, and poor health. Our findings suggest an inverse U-shaped relationship between alcohol consumption and wages with a peak at approximately 1.5 to 2.5 drinks per day on average. Thus, controlling for other variables and conditional on working, moderate alcohol users have higher wages than abstainers and heavy drinkers at these worksites. We also find evidence that alcohol use is related to wages through human capital variables.

Adult↗
PubMed1990

Comments on a time-dependent version of the linear-quadratic model.

The accuracy and interpretation of the "LQ + time" model (E = D(alpha + beta d) - gamma T) are discussed. Evidence is presented, based on data in the literature, that this model does not accurately describe the changes in isoeffect dose occurring with protraction of the overall treatment time during fractionated irradiation of the lung. This lack of fit of the model explains, in part, the surprisingly large values of gamma/alpha that have been derived from experimental lung data. The large apparent time factors for lung suggested by the model are also partly explained by the fact that gamma T/alpha, despite having units of dose, actually measures the influence of treatment time on the effect scale, not the dose scale, and is shown to consistently overestimate the change in total dose. The unusually high values of alpha/beta that have been derived for lung using the model (approximately 5 Gy) are shown to be influenced by the method by which the model was fitted to data. Reanalyses of the data using a more statistically valid regression procedure produce estimates of alpha/beta more typical of those usually cited for lung (approximately 3 Gy). Most importantly, published isoeffect data from lung indicate that the true deviation from the linear-quadratic (LQ) model is nonlinear in time, instead of linear, and also depends on other factors such as the effect level and the size of dose per fraction. Thus, we do not advocate the use of the "LQ + time" expression as a general isoeffect model.

Animals↗
PubMed1996

Gap-43 message levels in anterior cerebellum in Alzheimer's disease.

We have previously reported that decreased growth-associated protein (GAP-43) message in frontal association cortex (area 9) of Alzheimer's disease (AD) patients is associated with increased density of neurons containing neurofibrillary tangles (NFTs) [9]. This finding leads to the hypothesis that decreased GAP-43 message in AD may be related to NFTs, rather than to some other aspect of AD pathology. Therefore, we predicted that in areas of brain unaffected by NFTs in AD the GAP-43 message levels should be similar to those of controls. The cerebellum is known to have a number of pathologies of AD, including diffuse plaques (DPs), microglial activation and reactive astrocytes. NFTs, however, are not typically found in the cerebellum. mRNA was extracted from anterior cerebellum of AD and control cases, Northern- and slot-blotted and hybridized against a GAP-43 probe. Poly(dT) and glucose-3-phosphate dehydrogenase probes were used for normalization. The average relative GAP-43 message level was 0.582 in the AD cases and 0.448 in control cases. This 23% difference failed to reach statistical significance. Regression analysis within the AD group demonstrated that GAP-43 message level in cerebellar cortex was not significantly correlated with diffuse plaque density in cerebellar cortex. GAP-43 message levels in cerebellar cortex were also not correlated with summed density of neuritic plaques or summed density of NFTs in cortical regions-here used as an index of severity of disease. The data reported here also emphasize that the (NFT-dependent) reduction in GAP-43 mRNA levels previously reported in frontal association cortex in Alzheimer's disease [9] appears to be region specific and not a general brain phenomenon. The preservation of normal GAP-43 message levels in the cerebellum in AD is consistent with the hypothesis that events related to NFT formation have a major impact on the expression of GAP-43 in Alzheimer's disease.

Alzheimer Disease↗
PubMed1991

Covariate adjustment of treatment effects in clinical trials.

It has been previously shown that the magnitude of adjustment of a treatment effect Z value for a baseline covariate depends on both the degree of disparity of the covariate between the treatment groups and the degree of correlation between the covariate and the outcome variable. In this article, the magnitude of adjustment is further explored and tabulated for a variety of situations. Theoretically, the magnitude of adjustment may be great, for example, from Z = 3.0 unadjusted to 13.1 adjusted, or Z = 2.0 unadjusted to -1.6 adjusted. However, in more practically realistic situations the magnitude of adjustment will generally fall in the range of 1 to 2 standard error units, when the Z value for disparity of the covariate between the treatment groups is as great as +/- 3. If there is perfect comparability of treatment groups with respect to baseline covariate, there may still be a sizable difference between adjusted and unadjusted Z value for treatment effect, with adjusted almost always greater in absolute value than unadjusted Z value.

Clinical Trials as Topic↗
PubMed1989

Statistical analysis of complex health and social data.

The selection of appropriate statistical models is a vital aspect of the analysis of complex health and social data. This paper examines issues related to model selection. It is concluded that the range of methods now available makes old controversies less relevant. Researchers may now look for strategies that combine the advantages of non-parametric and parametric approaches without automatically accepting the limitations of either. Examples of new directions are provided.

Analysis of Variance↗
PubMed1994

Health insurance coverage among the elderly.

The research reported in this paper examines the decision to have private health insurance by elderly Medicare enrollees. Models allowing both stimultaneity and a joint error structure between health insurance and use of medical care are considered. We find that common unobserved variables underlying the joint errors are important determinants in the decision to purchase private health insurance. Simultaneity is present only between the decision to have private health insurance and the probability of visiting a doctor. Health status and functional limitations are important determinants of the decision to have private health insurance in addition to Medicare coverage. Other personal characteristics (age, sex, race and education), as well as household income, Medicaid enrollment, and the employment of a family member are also found to be related to the decision to have private health insurance.

Activities of Daily Living↗
PubMed1989

Doubts on the adequacy of the principal component varimax analysis for the identification of event-related brain potential components: a commentary on Glaser and Ruchkin, and Donchin and Heffley.

Glaser and Ruchkin (1976) as well as Donchin and Heffley (1978) have proposed principal component varimax analysis (PCVA) for identification of event-related potential (ERP) components. In the present commentary, the assumptions on the ERP data structure of this procedure are considered. They are formalized and compared with a simpler and more restrictive hypothesized ERP data structure containing autoregressive properties. Referring to the principle of parsimony, the comparison of the two models shows the superiority of the hypothesized autoregressive structure: to fit a set of ERP data equally well, the autoregressive structure does so with considerably more parsimonious assumptions, i.e. with fewer parameters to be estimated, than the structure induced by PCVA. It is concluded that an appropriate model for the ERP data structure should take the autocorrelated nature of the ERP into account.

Contingent Negative Variation↗
PubMed1989

Doubting these doubts--a reply to Collet.

In reply to Collet (1989) it is argued that principal component analysis (PCA) of event-related potentials is not invalidated or disproved by the arguments of his comment. The main points of this reply are as follows. First, since Collet's analysis was based on the correlation matrix only, it cannot disprove assumptions of PCA which do not constrain the correlation (or covariance) matrix. Second, his model utilizes results of PCA which invalidate the comparison and parameter count. Third, his model does not allow specific predictions and does not lead to real data reduction, which reverses Collet's argument based on the principle of parsimony. Fourth, his model is less general than PCA, as it could apply (if at all) to slow brain potentials only.

Evoked Potentials↗
PubMed1989

On why Collet's doubts regarding the PCA are misplaced.

Collet's (1989) doubts regarding the efficacy of principal component analysis (PCA) as a tool in the study of event-related brain potentials (ERPs) are unpersuasive. The substantive point reported by Collet is that data points are autocorrelated over the time period during which a single, PCA-defined, ERP component predominates. Such autocorrelation has long been recognized by investigators in the field, and its confirmation by autoregressive modeling does not provide useful information about issues central to ERP data analysis. Furthermore, because he fails to take a full count of the number of parameters used in his autoregressive model his argument from parsimony is flawed. In any event, Collet's argument misperceives the heuristic mode in which PCA is used in actual studies.

Evoked Potentials↗