Multilevel analysis for a longitudinal clinical trial

This study applies the multilevel analysis technique to longitudinal data of a large clinical trial. The technique accounts for the correlation at different levels when modeling repeated blood pressure measurements taken throughout the trial. This modeling allows for closer inspection of the remaining correlation and non-homogeneity of variance in the data. Three methods of modeling the correlation were compared. ^

An ordinal logistic regression model with misclassification of the outcome variable and categorical covariate

Ordinal outcomes are frequently employed in diagnosis and clinical trials. Clinical trials of Alzheimer's disease (AD) treatments are a case in point using the status of mild, moderate or severe disease as outcome measures. As in many other outcome oriented studies, the disease status may be misclassified. This study estimates the extent of misclassification in an ordinal outcome such as disease status. Also, this study estimates the extent of misclassification of a predictor variable such as genotype status. An ordinal logistic regression model is commonly used to model the relationship between disease status, the effect of treatment, and other predictive factors. A simulation study was done. First, data based on a set of hypothetical parameters and hypothetical rates of misclassification was created. Next, the maximum likelihood method was employed to generate likelihood equations accounting for misclassification. The Nelder-Mead Simplex method was used to solve for the misclassification and model parameters. Finally, this method was applied to an AD dataset to detect the amount of misclassification present. The estimates of the ordinal regression model parameters were close to the hypothetical parameters. β1 was hypothesized at 0.50 and the mean estimate was 0.488, β2 was hypothesized at 0.04 and the mean of the estimates was 0.04. Although the estimates for the rates of misclassification of X1 were not as close as β1 and β2, they validate this method. X 1 0-1 misclassification was hypothesized as 2.98% and the mean of the simulated estimates was 1.54% and, in the best case, the misclassification of k from high to medium was hypothesized at 4.87% and had a sample mean of 3.62%. In the AD dataset, the estimate for the odds ratio of X 1 of having both copies of the APOE 4 allele changed from an estimate of 1.377 to an estimate 1.418, demonstrating that the estimates of the odds ratio changed when the analysis includes adjustment for misclassification. ^

THE BACON CHOW STUDY: THE RELATIONSHIP BETWEEN INFECTIOUS ILLNESS AND GROWTH OF INFANTS

Multiple dietary deficiencies and high rates of infectious illness are major health problems leading to malnutrition and limitation of growth of children in developing countries. Longitudinal studies which provide information on illness incidence and growth velocity are needed in order to untangle the complex interrelationship between nutrition, illness and growth. From 1967 to 1973, researchers led by Dr. Bacon Chow of the Johns Hopkins University School of Hygiene undertook a quasi-experimental prospective study in Suilin Township, Taiwan to determine the effects of a nutritional supplement to the diets of pregnant and lactating women on the growth, development and resistance to disease of their offspring. This dissertation presents results from the analysis of infant morbidity and postnatal growth.^ Maternal nutritional supplementation has no apparent effect on the postnatal growth or morbidity of infants. Significant sex differences exist in growth response to illness and in illness susceptibility. Male infants have more diarrhea and upper respiratory illness. Respiratory illness is positively associated with growth rate in weight in the first semester of life. Diarrhea is significantly negatively associated with growth in length in the second semester. Small-for-date infants are more susceptible to illness in general and have a different pattern of growth response than large-for-date infants.^ Principal components analysis of illness data is shown to be an effective technique for making more precise use of ambiguous morbidity data. Multiple regression with component scores is an accurate method for estimating variance in growth rate predicted by indepenent illness variables. A model is advanced in which initial postnatal growth rate determines subsequent susceptibility to nutritional stress and infection. Initial growth rate is a function of prenatal nutrition, but is not significantly affected by maternal supplementation during gestation or lactation. Critical evaluation is made of nutritional supplementation programs which do not afford disease control.^