Multiple imputation for missing continuous and dichotomous data: The effects of patterns of missingness and variable correlations on type I error with small samples

The purpose of this study is to investigate the effects of predictor variable correlations and patterns of missingness with dichotomous and/or continuous data in small samples when missing data is multiply imputed. Missing data of predictor variables is multiply imputed under three different multivariate models: the multivariate normal model for continuous data, the multinomial model for dichotomous data and the general location model for mixed dichotomous and continuous data. Subsequent to the multiple imputation process, Type I error rates of the regression coefficients obtained with logistic regression analysis are estimated under various conditions of correlation structure, sample size, type of data and patterns of missing data. The distributional properties of average mean, variance and correlations among the predictor variables are assessed after the multiple imputation process. ^ For continuous predictor data under the multivariate normal model, Type I error rates are generally within the nominal values with samples of size n = 100. Smaller samples of size n = 50 resulted in more conservative estimates (i.e., lower than the nominal value). Correlation and variance estimates of the original data are retained after multiple imputation with less than 50% missing continuous predictor data. For dichotomous predictor data under the multinomial model, Type I error rates are generally conservative, which in part is due to the sparseness of the data. The correlation structure for the predictor variables is not well retained on multiply-imputed data from small samples with more than 50% missing data with this model. For mixed continuous and dichotomous predictor data, the results are similar to those found under the multivariate normal model for continuous data and under the multinomial model for dichotomous data. With all data types, a fully-observed variable included with variables subject to missingness in the multiple imputation process and subsequent statistical analysis provided liberal (larger than nominal values) Type I error rates under a specific pattern of missing data. It is suggested that future studies focus on the effects of multiple imputation in multivariate settings with more realistic data characteristics and a variety of multivariate analyses, assessing both Type I error and power. ^

Indeterminate QunatiFERON-TB Gold In-Tube assay results in association with procedural specimen collection in children

BACKGROUND. The development of interferon-gamma release assays (IGRA) has introduced powerful tools in diagnosing latent tuberculosis infection (LTBI) and may play a critical role in the future of tuberculosis diagnosis. However, there have been reports of high indeterminate results in young patient populations (0-18 years). This study investigated results of the QunatiFERON-TB Gold In-Tube (QFT-GIT) IGRA in a population of children (0-18 years) at Texas Children's Hospital in association with specimen collection procedures using surrogate variables. ^ METHODS. A retrospective case-control study design was used for this investigation. Cases were defined as having QFT-GIT indeterminate results. Controls were defined as having either positive or negative results (determinates). Patients' admission status, staff performing specimen collection, and specific nurse performing specimen collection were used as surrogates to measure specimen collection procedures. ^ To minimize potential confounding, abstraction of patients' electronic medical records was performed. Abstracted data included patients' medications and evaluation at the time of QFT-GIT specimen collection in addition to their medical history. QFT-GIT related data was also abstracted. Cases and controls were characterized using chi-squared tests or Fisher's exact tests across categorical variables. Continuous variables were analyzed using one-way ANOVA and t-tests for continuous variables. A multivariate model was constructed by backward stepwise removal of statistically significant variables from univariate analysis. ^ RESULTS. Patient data was abstracted from 182 individuals aged 0-18 years from July 2010 to August 2011 at Texas Children's Hospital. 56 cases (indeterminates) and 126 controls (determinates) were enrolled. Cancer was found to be an effect modifier with subsequent stratification resulting in a cancer patient population too small to analyze (n=13). Subsequent analyses excluded these patients. ^ The exclusion of cancer patients resulted in a population of 169 patients with 49 indeterminates (28.99%) and 120 determinates (71.01%), with mean ages of 9.73 (95% CI: 8.03, 11.43) years and 11.66 (95% CI: 10.75, 12.56) years (p = 0.033), respectively. Median age of patients who were indeterminates and determinates were 12.37 and 12.87 years, respectively. Lack of data for our specific nurse surrogate (QFTNurse) resulted in its exclusion from analysis. The final model included only our remaining surrogate variables (QFTStaff and QFTInpatientOutpatient). The staff collecting surrogate (QFTStaff) was found to be modestly associated with indeterminates when nurses collected the specimen (OR = 1.54, 95% CI: 0.51, 4.64, p = 0.439) in the final model. Inpatients were found to have a strong and statistically significant association with indeterminates (OR = 11.65, 95% CI: 3.89, 34.9, p < 0.001) in the final model. ^ CONCLUSION. Inpatient status was used as a surrogate for indication of nurse drawn blood specimens. Nurses have had little to no training regarding shaking of tubes versus phlebotomists regarding QFT-GIT testing procedures. This was also measured by two other surrogates; specifically a medical note stating whether a nurse or phlebotomist collected the specimen (QFTStaff) and the name and title of the specific nurse if collection was performed by a nurse (QFTNurse). Results indicated that inpatient status was a strong and statistically significant factor for indeterminates, however, nurse collected specimens and indeterminate results had no statistically significant association in non-cancer patients. The lack of data denoting the specific nurse performing specimen collection excluded the QFTNurse surrogate in our analysis. ^ Findings suggests training of staff personnel in specimen procedures may have little effect on the number of indeterminates while inpatient status and thus possibly illness severity may be the most important factor for indeterminate results in this population. The lack of congruence between our surrogate measures may imply that our inpatient surrogate gauged illness severity rather than collection procedures as intended. ^ Despite the lack of clear findings, our analysis indicated that more than half of indeterminates were found in specimens drawn by nurses and as such staff training may be explored. Future studies may explore methods in measuring modifiable variables during pre-analytical QFT-GIT procedures that can be discerned and controlled. Identification of such measures may provide insight into ways to lowering indeterminate QFT-GIT rates in children.^

Alpha C protein of group B Streptococcus as a potential vaccine candidate

Group B Streptococcus (GBS) is a leading cause of life-threatening infection in neonates and young infants, pregnant women, and non-pregnant adults with underlying medical conditions. Immunization has theoretical potential to prevent significant morbidity and mortality from GBS disease. Alpha C protein (α C), found in 70% of non-type III capsule polysaccharide group B Streptococcus, elicits antibodies protective against α C-expressing strains in experimental animals and is an appealing carrier for a GBS conjugate vaccine. We determined whether natural exposure to α C elicits antibodies in women and if high maternal α C-specific serum antibody at delivery is associated with protection against neonatal disease. An ELISA was designed to measure α C-specific IgM and IgG in human sera. A case-control design (1:3 ratio) was used to match α C-expressing GBS colonized and non-colonized women by age and compare quantified serum α C-specific IgM and IgG. Sera also were analyzed from bacteremic neonates and their mothers and from women with invasive GBS disease. Antibody concentrations were compared using t-tests on log-transformed data. Geometric mean concentrations of α C-specific IgM and IgG were similar in sera from 58 α C strain colonized and 174 age-matched non-colonized women (IgG 245 and 313 ng/ml; IgM 257 and 229 ng/ml, respectively). Delivery sera from mothers of 42 neonates with GBS α C sepsis had similar concentrations of α C-specific IgM (245 ng/ml) and IgG (371 ng/ml), but acute sera from 13 women with invasive α C-expressing GBS infection had significantly higher concentrations (IgM 383 and IgG 476 ng/ml [p=0.036 and 0.038, respectively]). Convalescent sera from 5 of these women 16-49 days later had high α C-specific IgM and IgG concentrations (1355 and 4173 ng/ml, respectively). In vitro killing of α C-expressing GBS correlated with total α C-specific antibody concentration. Invasive disease but not colonization elicits α C-specific IgM and IgG in adults. Whether α C-specific IgG induced by vaccine would protect against disease in neonates merits further investigation. ^

A SIMULATION STUDY OF THE SIZE AND POWER OF SOME TWO-SAMPLE TESTS FOR UNEQUALLY CENSORED SURVIVAL DATA (RANK TESTS, LOGRANK, PROPORTIONAL HAZARDS, WILCOXON TESTS, EXPONENTIAL DISTRIBUTION)

Sizes and power of selected two-sample tests of the equality of survival distributions are compared by simulation for small samples from unequally, randomly-censored exponential distributions. The tests investigated include parametric tests (F, Score, Likelihood, Asymptotic), logrank tests (Mantel, Peto-Peto), and Wilcoxon-Type tests (Gehan, Prentice). Equal sized samples, n = 18, 16, 32 with 1000 (size) and 500 (power) simulation trials, are compared for 16 combinations of the censoring proportions 0%, 20%, 40%, and 60%. For n = 8 and 16, the Asymptotic, Peto-Peto, and Wilcoxon tests perform at nominal 5% size expectations, but the F, Score and Mantel tests exceeded 5% size confidence limits for 1/3 of the censoring combinations. For n = 32, all tests showed proper size, with the Peto-Peto test most conservative in the presence of unequal censoring. Powers of all tests are compared for exponential hazard ratios of 1.4 and 2.0. There is little difference in power characteristics of the tests within the classes of tests considered. The Mantel test showed 90% to 95% power efficiency relative to parametric tests. Wilcoxon-type tests have the lowest relative power but are robust to differential censoring patterns. A modified Peto-Peto test shows power comparable to the Mantel test. For n = 32, a specific Weibull-exponential comparison of crossing survival curves suggests that the relative powers of logrank and Wilcoxon-type tests are dependent on the scale parameter of the Weibull distribution. Wilcoxon-type tests appear more powerful than logrank tests in the case of late-crossing and less powerful for early-crossing survival curves. Guidelines for the appropriate selection of two-sample tests are given. ^

Standardization of data collection for the Lifetime Surveillance of Astronaut Health (LSAH)

Problems due to the lack of data standardization and data management have lead to work inefficiencies for the staff working with the vision data for the Lifetime Surveillance of Astronaut Health. Data has been collected over 50 years in a variety of manners and then entered into a software. The lack of communication between the electronic health record (EHR) form designer, epidemiologists, and optometrists has led to some level to confusion on the capability of the EHR system and how its forms can be designed to fit all the needs of the relevant parties. EHR form customizations or form redesigns were found to be critical for using NASA's EHR system in the most beneficial way for its patients, optometrists, and epidemiologists. In order to implement a protocol, data being collected was examined to find the differences in data collection methods. Changes were implemented through the establishment of a process improvement team (PIT). Based on the findings of the PIT, suggestions have been made to improve the current EHR system. If the suggestions are implemented correctly, this will not only improve efficiency of the staff at NASA and its contractors, but set guidelines for changes in other forms such as the vision exam forms. Because NASA is at the forefront of such research and health surveillance the impact of this management change could have a drastic improvement on the collection of and adaptability of the EHR. Accurate data collection from this 50+ year study is ongoing and is going to help current and future generations understand the implications of space flight on human health. It is imperative that the vast amount of information is documented correctly.^

Mixture modeling for joint analysis of survival, discrete, and continuous data

Mixture modeling is commonly used to model categorical latent variables that represent subpopulations in which population membership is unknown but can be inferred from the data. In relatively recent years, the potential of finite mixture models has been applied in time-to-event data. However, the commonly used survival mixture model assumes that the effects of the covariates involved in failure times differ across latent classes, but the covariate distribution is homogeneous. The aim of this dissertation is to develop a method to examine time-to-event data in the presence of unobserved heterogeneity under a framework of mixture modeling. A joint model is developed to incorporate the latent survival trajectory along with the observed information for the joint analysis of a time-to-event variable, its discrete and continuous covariates, and a latent class variable. It is assumed that the effects of covariates on survival times and the distribution of covariates vary across different latent classes. The unobservable survival trajectories are identified through estimating the probability that a subject belongs to a particular class based on observed information. We applied this method to a Hodgkin lymphoma study with long-term follow-up and observed four distinct latent classes in terms of long-term survival and distributions of prognostic factors. Our results from simulation studies and from the Hodgkin lymphoma study demonstrated the superiority of our joint model compared with the conventional survival model. This flexible inference method provides more accurate estimation and accommodates unobservable heterogeneity among individuals while taking involved interactions between covariates into consideration.^

Longitudinal categorical data analysis: A continuous -time Markov chain approach

In this dissertation, we propose a continuous-time Markov chain model to examine the longitudinal data that have three categories in the outcome variable. The advantage of this model is that it permits a different number of measurements for each subject and the duration between two consecutive time points of measurements can be irregular. Using the maximum likelihood principle, we can estimate the transition probability between two time points. By using the information provided by the independent variables, this model can also estimate the transition probability for each subject. The Monte Carlo simulation method will be used to investigate the goodness of model fitting compared with that obtained from other models. A public health example will be used to demonstrate the application of this method. ^