Biomedical subjects

W L May

Publications and source records attributed to W L May.

30 records · Page 2Linked to original sources

PubMed1997

Confidence intervals for differences in correlated binary proportions.

An experiment to assess the efficacy of a particular treatment or process often produces dichotomous responses, either favourable or unfavourable. When we administer the treatment on two occasions to the same subjects, we often use McNemar's test to investigate the hypothesis of no difference in the proportions on the two occasions, that is, the hypothesis of marginal homogeneity. A disadvantage in using McNemar's statistic is that we estimate the variance of the sample difference under the restriction that the marginal proportions are equal. A competitor to McNemar's statistic is a Wald statistic that uses an unrestricted estimator of the variance. Because the Wald statistic tends to reject too often in small samples, we investigate an adjusted form that is useful for constructing confidence intervals. Quesenberry and Hurst and Goodman discussed methods of construction that we adapt for constructing confidence intervals for the differences in correlated proportions. We empirically compare the coverage probabilities and average interval lengths for the competing methods through simulation and give recommendations based on the simulation results.

Analysis of Variance↗

PubMed1997

The validity and power of tests for equality of two correlated proportions.

With measurements taken on subjects over time, on matched pairs of subjects or on clusters of subjects, the data often contain pairs of correlated dichotomous responses. McNemar's test is perhaps the best known test to compare two correlated binomial proportions. The salient feature of McNemar's test is that we compute the variance of the contrast estimator under the restriction that the null hypothesis is true. Wald's test, on the other hand, does not require that restriction. As a consequence, Wald's statistic is always greater in magnitude than McNemar's statistic when the marginal proportions are unequal, but there is a problem with the validity of both McNemar's test and Wald's test with small to moderate samples. There have been various modifications suggested for McNemar's test to improve its performance. We propose a modified Wald's test that is valid in small to moderate samples and maintains good power. We also evaluate the performance of McNemar's test and Wald's test with and without modifications to enhance validity as well as the performance of the large sample likelihood ratio test and an exact test of the equality of correlated binomial proportions. In a smaller study, we compare the behaviour of a test based on the James-Stein estimator of the common odds ratio proposed by Liang and Zeger to McNemar's test and Wald's test.

Child↗

PubMed1997

A SAS macro for constructing simultaneous confidence intervals for multinomial proportions.

Quesenberry and Hurst (1964), Goodman (1965) and Fitzpatrick and Scott (1987) proposed simultaneous construction of confidence intervals for multinomial proportions, however. statistical computing packages do not generally give one the option of specifying the type of construction to be used. We have written a SAS macro using PROC IML that takes multinomial cell counts as input and returns simultaneous confidence intervals with the user-specified coverage probability. Two main features of the macro are its ease of use and its flexibility in allowing the user to choose among six methods of constructing confidence intervals for multinomial proportions. Based on simulation May and Johnson (1997) recommended the intervals proposed by Goodman (1965) in most practical applications.

Binomial Distribution↗

PubMed1996

Generalized estimating equations for multivariate response with the variates having different distributions.

This paper addresses the problem of analyzing multivariate response with the variates having different distributions. We use the generalized estimating equations proposed by Prentice and Zhao (1) to estimate mean and covariance parameters. Wald statistics are used to test hypotheses about the parameters. Data from a two-group study of competing programs to increase memory recall in geriatric patients are analyzed. We are interested in determining whether there is a simultaneous increase in test scores with one variate being continuous and another being dichotomous. In a second example, comparisons of before-and-after treatment serum cholesterol levels and number of snacks eaten per day are made in an experiment to determine efficacy of a program aimed at simultaneously reducing both measurements.

Cholesterol↗

PubMed1995

Combining 2 x 2 tables that contain structural zeros.

When one cell or more of a contingency table is necessarily zero, the table is said to contain structural zeros. These structural voids may be inherent to the problem at hand, or, in some applications, they may be introduced intentionally through the experimental design. With data classified by a third stratification variable, one needs a method for analysing a set of independent 2 x 2 tables. We propose Cochran-Mantel-Haenszel type analyses that combine observations across strata. We also propose tests of hypotheses pertaining to the marginal and conditional probabilities.

Data Interpretation, Statistical↗

PubMed1995

Some applications of the analysis of multivariate normal data with missing observations.

Methods are proposed for the analysis of data from a multivariate normal distribution when observations are missing completely at random on some of the variates. Park (1) proved the equivalence of the solutions given by maximum likelihood and generalized estimating equations when data are complete and an unstructured covariance is assumed. He suggested that generalized estimating equations may be used if sample sizes are large relative to the amount of missing data and the estimated covariance matrix is positive definite. We give several examples indicating that the estimating equations give results similar to those of maximum likelihood when smoothing of the covariance matrix to eliminate nonpositive definiteness is not encountered as, for example, under an assumption of exchangeable correlation. Generalized linear models are formulated that are appropriate for a wide class of experimental plans.

Blood Coagulation↗

PubMed1994

Effect of out-of-hospital albuterol inhalation treatments on patient comfort and morbidity.

STUDY OBJECTIVE: To determine the effect of the institution of out-of-hospital albuterol treatments for acute asthma on out-of-hospital time intervals emergency department (ED) morbidity. DESIGN: Retrospective chart and ambulance run report review. PARTICIPANTS: All patients who presented with a diagnosis of asthma on the ED record and ambulance run report during 1988 and 1990. RESULTS: Ninety patients treated prior to the routine use of nebulized albuterol (in 1988) and 86 patients treated during the period after the institution of nebulized albuterol (in 1990) were compared. Age, sex, scene and travel times, vital signs, peak expiratory flow rates, length of ED stay, hospitalizations, ED treatment, and mortality were recorded for each patient. Changes over time were determined using a cohort of asthma patients in each year as controls. Severity was recorded as low (less than 3) or high (3 or more) on a modified Fischl index for asthma. Groups were compared by F test, chi 2, or t test with P < .05 considered significant. Because there were differences for admission rates, first-hospital peak expiratory flow rates, and aminophylline use in control groups, the experimental groups could not be compared. No significant difference in travel interval, steroid use, albuterol treatments, or length of ED stay was determined in either the experimental or control group. Scene time (9.7 minutes versus 12.1 minutes) was significantly shorter in 1988 than in 1990 (difference, 2.4 minutes; 95% confidence interval, 0.2 to 4.6; P < .01). The percentage of patients with high initial severity of disease (85% versus 66%; P < .001) and the percentage of patients with first posttreatment peak expiratory flow rates of less than 120 L/min (77% versus 58%; P < .001) were significantly higher in 1988 than in 1990. There was no significant difference in scene time, severity scores, or first posttreatment peak expiratory flow rates in control groups. CONCLUSION: The institution of out-of-hospital use of nebulized albuterol increased scene time and increased first posttreatment peak expiratory flow rates significantly. There was a significant decrease in initial patient severity as measured on the modified Fischl index, although the validity of this index has not been established. It did not affect travel interval, length of stay in the ED, or medication use after ED presentation.

Acute Disease↗

PubMed1993

A computer package for the multivariate nonparametric rank test in completely randomized experimental designs.

When research data are measured on at least an ordinal scale and the assumptions required in the theory underlying parametric statistical methods are in question, nonparameteric procedures based on the ranks provide a sound approach to statistical analysis. Biomedical investigations, especially clinical trials, typically involve multivariate response and therefore multivariate statistical methods are called for in the interpretation of results. We discuss applications of the nonparametric multivariate rank test for completely randomized designs. Large sample theory can be used to support these statistical methods for assessing group differences in location. In small samples, randomization tests provide a basis for inferences. The execution of the procedure is facilitated by a computer program developed by the authors.

Algorithms↗

PubMedDate not supplied

Reduced placental perfusion causes an increase in maternal serum leptin.

OBJECTIVE: We tested the hypothesis that the inadequately perfused placenta increases production of leptin, which can be detected in maternal serum. STUDY DESIGN: Sprague-Dawley rats (n=13), on day 14 of gestation, had placement of clips on the aorta and the ovarian arteries providing 35 per cent occlusion of the vessels. Eight rats had sham surgery and 14 rats served as non-surgical controls. All animals were sacrificed on day 19 of gestation. Maternal serum was obtained, and pups and placentae were weighed. RESULTS: Both placental weights and pup weights were reduced due to reduced uterine perfusion and were negatively correlated with maternal serum leptin (P=0.018 and 0.028, respectively). Maternal serum leptin was increased in the treatment group (2.21 ng/ml+/-64 ng/ml) compared to controls (1.66 ng/ml+/-38 ng/ml) (P=0.031). CONCLUSIONS: Our findings suggest that reduced placental perfusion results in an increase in maternal serum leptin. Further investigation is needed to determine if maternal serum leptin may be useful in identifying pregnancies with uteroplacental insufficiency.

Animals↗

PubMedDate not supplied

Combined analysis of cytokine genotype polymorphism and the level of expression with allograft function in African-American renal transplant patients.

Cytokine gene polymorphism and expression levels were evaluated in a group of African-American patients who had undergone renal transplantation. It was hypothesized that possession of specific cytokine alleles might be influential in predisposing the recipient to allograft rejection. Thus, we sought to establish a relationship between cytokine gene polymorphism, the levels of cytokine expression, and the outcome of allograft function. Cytokine genotypes and mRNA transcript levels of IL-2, TNF-alpha, TGF-beta1, IL-10, IL-6 and IFN-gamma were determined using peripheral blood cells. Genomic DNA samples from 77 transplant recipients and 77 controls were tested by a multiplex PCR with specific primers for the above cytokines. The frequency distributions of cytokines were analyzed in respect to the clinical characterization, including delayed graft function (DGF), rejection episodes (REs) and stable graft function (SGF). The mRNA transcript level was tested both at pre- and early post-transplantation (day 1 and day 4) with primers for coding regions of the above cytokines in a RT-PCR assay. The majority of recipients with successful graft function were matched with their donors for only three out of the six HLA alleles. We have shown that the TGF-beta1 T/C G/G high producer and IFN-gamma T/A intermediate producer genotypes were associated with allograft rejection, whereas low IFN-gamma producer and high IL-10 producer genotypes were significantly protective of the allograft. There was some correlation between the TGF-beta1 high producer genotype and DGF, but it was not statistically significant. Overall, 77% of those who experienced REs carried the TGF-beta1 T/C G/G, high producer genotype as compared with 52% who experienced DGF, 39% with SGF (P<0.01, RR=2.0), and 27.3% of controls (P<0.003, RR=2.6). The IFN-gamma T/A intermediate producer genotype was found in 69.2% of patients with REs as compared with 26.8% of patients with SGF (P<0.008, RR=2.85). The IL-10, ATA/ATA low producer genotype was found in 38.5% of recipients with REs and 14.6% of recipients without REs (P<0.04, RR=0.53). Expression levels of mRNA transcript were correlated with genotype data, except for the TGF-beta1 high producer genotype where there was no significant difference between the level of mRNA transcript at pre- and post-transplantation. Low DRbeta1 and high DPbeta1 expression by recipient peripheral blood mononuclear cells before transplantation was associated with more SGF, whereas high DRbeta1 and low DPbeta1 expression at pretransplantation was associated with more REs (DRbeta1, P<0.001 and DPbeta1, P<0.05, respectively). We concluded that, dual analysis of cytokine genotype and expression levels by peripheral cells may be an important clue to understanding the contribution of the recipient's immune response to an allograft pre- and post-transplantation. Identification of peripheral markers diagnostic of rejection could allow advance anticipation of clinical outcome, and might reduce the need for tissue biopsy.

Black or African American↗

PubMedDate not supplied

Symmetry in square contingency tables: tests of hypotheses and confidence interval construction.

Bowker's test, a generalization of McNemar's test, performs well under the hypothesis of symmetry, but the estimator of variance used in the test is biased when the table is asymmetric and this calls into question the test's performance in non-null situations. We seek an alternative to Bowker's test in search of methods for simultaneous inference that are valid when the hypothesis of symmetry is false. We apply multivariate normal theory to develop chi-square tests and simultaneous confidence intervals for inferences concerning symmetry in k X k contingency tables. We propose a modified Wald statistic as a competitor to Bowker's test. We also proffer quadratic estimators of confidence intervals. In large samples, the recommended test statistic rejects the null hypothesis at the stated level of significance when the null hypothesis is true and always rejects with greater power than Bowker's test. The proffered interval estimators provide good simultaneous coverage of the pairwise differences between the population proportions at the stated confidence level.

Biometry↗

PubMedDate not supplied

Relationship between serum inositol concentration and development of retinopathy of prematurity: a prospective study.

PURPOSE: To examine the relationship between the intake of sugar inositol, serum inositol levels, and ROP in three groups of low birthweight infants receiving feedings containing various concentrations of inositol. METHODS: Infants with a birthweight <1500 g, with severe lung disease, were eligible for the study when they began enteral feedings. Infant formulas contained three different inositol concentrations: 2500, 710, and 242 micromol/L. Serum inositol concentrations were averaged over specific time intervals. A logistic regression model was used to investigate the confounding effect of duration of mechanical ventilation and oxygen therapy, birthweight, Apgar score, and serum inositol concentration on development of ROP. RESULTS: Infants receiving high inositol formula and with higher serum inositol concentrations at birth and after 30 days had a statistically significant lower incidence of severe ROP than those receiving the lower inositol formula and with lower serum concentrations (P<.05). The effective serum inositol concentration (EC90) associated with lesser disease was >215 micromol/L. By logistic regression, the odds of developing severe ROP were greater among infants with low serum inositol concentration (odds ratio=4.7, 95% confidence interval 0.90-24.8, P=.017). CONCLUSION: Inositol supplementation may help prevent the most severe form of ROP.

Dietary Carbohydrates↗