Spatio-temporal interaction with disease mapping.

Markov chain Monte Carlo methods are used to estimate mortality rates under a Bayesian hierarchical model. Spatial correlations are introduced to examine spatial effects relative to both regional and regional changes over time by groups. A special feature of the models is the inclusion of longitudinal variables which will describe temporal trends in mortality or incidences for different population groups. Disease maps are used to illustrate the role of different parameters in the model and pinpointing areas of interesting patterns. The methods are demonstrated by male cancer mortality data from the state of Missouri during 1973-1992. Of special interest will be the geographic variations in the trend of lung cancer mortality over the recent past. Marginal posterior distributions are used to examine effects due to spatial correlations and age difference in temporal trends. Numerical results from the Missouri data show that although spatial correlations exist, they do not have a large effect on the estimated mortality rates.

Bayesian design for dose-response curves with penalized risk.

This paper considers some Bayesian design problems in quantal response analysis. An experimenter must choose a set of dose levels and number of independent observations to take at these levels, subject to a total sample size, in order to estimate some characteristic phi, e.g., ED50, of a tolerance distribution F theta, where theta is the vector of unknown parameters. It is shown for the logistic model that the sampling variability of the posterior variance of phi is such that the predicted posterior variance alone is an undesirable criterion for design selection. A family of penalty functions is introduced that penalizes any excess in the posterior variance over the expected or predicted variance and protects against unexpected outcomes. The goal is to find a design that avoids experimental results with little information, at the expense of a small sacrifice in the Bayes risk. Numerical results indicate that the chance of an extreme posterior variance can be reduced by sacrificing a small amount of posterior risk.

Analysis of mortality rates via marginal extended quasi-likelihood.

We use a mixed Poisson regression model with extra variation to analyse mortality data cross-classified by age and geographic region. We use estimates of dispersion parameter and fixed effects parameters, obtained by maximizing a marginal quasi-likelihood function, to estimate mortality rates in an empirical Bayes manner. This is a modification of an earlier method by Tsutakawa that used the likelihood function and is computationally impractical for routine use. In order to apply the extended quasi-likelihood function, a linear variance function (of the mean) is used in place of a quadratic one used previously. We illustrate the method with male lung cancer deaths in the state of Missouri.

Mixed model for analyzing geographic variability in mortality rates.

"A mixed model is proposed for the analysis of geographic variability in mortality rates. In addition to demographic parameters and random geographic parameters, the model includes additional random-effects parameters to adjust for extra-Poisson variability. The model uses a gamma-Poisson distribution with a random scale parameter having an inverse gamma prior. An empirical Bayes approach is used to estimate relative risks for geographic regions and annual rates for demographic groups within each region. Lung cancer in Missouri is used to motivate and illustrate the procedure."

Amiodarone and sustained ventricular arrhythmias: statistical evidence of drug effectiveness.

Previous studies have shown that amiodarone prevents sustained ventricular arrhythmias in 77% to 93% of patients. To date, a study using statistical analysis to verify the drug's effectiveness has not been reported. Amiodarone was given to 17 patients with drug refractory sustained ventricular arrhythmias. All patients had serious underlying heart disease including coronary artery disease (15 patients) or cardiomyopathy (two patients). Ten patients had angiographic evidence of a left ventricular aneurysm. All patients had left ventricular dysfunction. The mean left ventricular ejection fraction was 33%. In the 5.5 +/- 8.3 months prior to amiodarone, these 17 patients had documented sustained ventricular arrhythmias requiring countershock (41 episodes), overdrive pacing (four episodes), or intravenous drugs (three episodes). Amiodarone was given as a loading dose (1 gm/day for 10 days) and a maintenance dose (200 to 600 mg/day). During a follow-up period of 8.9 +/- 5.7 months, only eight episodes occurred requiring countershock (5) or overdrive pacing (2); one patient died suddenly. A statistical test constructed for this problem showed a significant (p greater than 0.001) reduced risk of experiencing a sustained ventricular arrhythmia after amiodarone. This statistical model confirms previous studies showing that amiodarone prevents sustained ventricular arrhythmias and prevents sudden cardiac death.

Empirical Bayes estimation of cancer mortality rates.

Mortality rates for specific cancer types and age-sex groups computed for large numbers of cities typically show extreme fluctuation. This is primarily due to the rare occurrence of specific cancer deaths in most of the small and moderate size cities during a fixed time period. Assuming a Poisson death process, we use an empirical Bayes method to obtain adjusted rates that are more stable for comparison of cities and prediction of future mortality. We have chosen stomach and bladder cancers in Missouri cities to illustrate the problems, techniques and results.

PIP: An empirical Bayes method is used to obtain adjusted rates of cancer mortality that are more stable for use in comparisons among cities and to predict future mortality trends. The method is illustrated using data on stomach and bladder cancers in Missouri cities.

Estimation of cancer mortality rates: a Bayesian analysis of small frequencies.

A Bayesian method is presented for estimating mortality rates of specific diseases when the frequency of deaths over a specified time period is assumed to have a Poisson distribution with mean proportional to the population size. The estimators use information from related populations, each having its own rate which is assumed distributed according to a common prior distribution about which some information is available. The study was motivated by an epidemiological study on the geographic variation of cancer mortality in the state of Missouri. Data from this study are used to illustrate the method and to compare it to a somewhat simpler empirical Bayes method.

Influence of chlortetracycline feeding on salmonellosis in calves. I. Rate and duration of shedding. II. Severity of illness.

Studies were conducted to test the influence of the feeding of chlortetracycline (CTC) on the fecal shedding of S. typhimurium subsequent to experimentally induced infection in calves. Levels of 0, 20, 50, and 100 gm CTC per ton of feed were fed to groups of calves for a two-week period before inoculation and the resulting level of shedding of S. typhimurium quantified. At the 50 gm/ton level, the feeding of CTC was associated with a significantly higher level of shedding than in non-CTC fed controls, and the duration of shedding was longer. Calves fed at 50 and 100 gm CTC/ton were affected much more severely by the inoculation than calves receiving no CTC. The same was true to a lesser extent in the calves fed 20 gm/ton. Observations made on each calf included changes in body temperature, time of onset, severity and duration of diarrhea, straining, and anorexia. Since the fecal output of salmonellae is increased at the level of 50 gm/ton, the authors conclude that this commonly used level of CTC feeding in calves contributes to the size of the salmonella reservoir in nature, thus increasing the risk of exposure to both man and animals alike and otherwise complicates the problems of salmonellosis.

Statistical analysis of urinary delta-aminolevulinic acid (ALAU) excretion in the white-footed mouse associated with lead smelting.

ALAU in white-footed mice trapped in the vicinity of a lead smelter has been measured to study the biological effect of lead smelting operations and the rate at which the ALAU level diminishes after removing animals from contaminated environments. The statistical method which has been developed for this purpose shows that the ALAU levels are initially higher among animals near the smelter and diminish to a common level after 5-6 weeks. The statistical method is generally applicable to a variety of physiological functions which exhibit a decay curve with respect to time.

Therapeutic responses of piglets to experimentally induced colibacillosis.

The purpose of these experiments was to determine the effects of prior exposure to antimicrobials on subsequent treatment of artificially induced colibacillosis in pigs. One- to two-week-old piglets were given 10(7) multiple antibiotic resistant Escherichia coli orally. Two groups of the piglets received priming doses at different levels of the antimicrobial preparation, ASP-250 before challenge. The remaining group received no antimicrobial. Piglets ill as a result of the challenge were treated with chloramphenicol of chlortetracycline, or received no treatment. Chloramphenicol was significantly more effective than chlortetracycline in terminating colibacillosis in the primed and unprimed groups. There were fewer deaths and relapse cases in those groups which received chloramphenicol treatment. Results in piglets treated with chlortetracycline were not significantly better than those which received no treatment.

Effect of diet on fecal nutrients.

Calculated and chemically analyzed nutrient values for three weighed constant diets were dissimilar for calcium, magnesium, protein, and fat, and similar for fiber and ash. Fecal nutrient concentrations of ten healthy men fed the same constant diets were quite variable. Feeding constant weighed diets of similar composition does not assure the same fecal residue in the large colon of human subjects.

Quick test for comparing two populations with bivariate data.

A simple nonparametric test is given for testing the equality of two populations when the observations (X, Y) are bivariate and the alternative of interest is that the conditional c.d.f. of Y given X for one population dominates that for the other for every value of X. The procedure consists of fitting a curve through the combined set of data points and considering and conditional distribution of the number of points above the curve from one population, given certain information (depending on the design of the experiment), on the combined set of data. When used to compare two bivariate distributions, the test reduces to Fisher's exact test for 2 X 2 tables.

Effect of a high-beef diet on the fecal bacterial flora of humans.

Ten human volunteers completed a 4-month diet series consisting of 1 month each of a control diet, a meatless diet, a high-beef diet, and the same control diet. Fat and fiber contents were essentially the same in all four diets, but protein content was doubled during the high-beef diet. During the 4th week on each diet, three stool specimens collected from each volunteer were analyzed for chemical composition and content of facultative, aerobic, and anaerobic bacteria. The bacteriological data are presented in this paper. High beef protein consumption had little effect on the composition of the intestinal flora. There were no significant differences in total counts of facultative and aerobic or anaerobic organisms in the feces when volunteers were on meatless or high-beef diets. At the species level, when counts during the two control diets were comparable, in only three instances did the change from the meatless to a high-beef diet significantly influence the bacterial numbers. The ratio of mean counts of anaerobic to facultative and aerobic organisms was approximately 15:1 during the meatless diet and 34:1 during the high-meat diet. The data indicate that animal protein consumption has little effect on the fecal bacterial profile in humans.