An empirical evaluation of the Random Forests classifier models for variable selection in a large-scale lung cancer case-control study

Random Forests™ is reported to be one of the most accurate classification algorithms in complex data analysis. It shows excellent performance even when most predictors are noisy and the number of variables is much larger than the number of observations. In this thesis Random Forests was applied to a large-scale lung cancer case-control study. A novel way of automatically selecting prognostic factors was proposed. Also, synthetic positive control was used to validate Random Forests method. Throughout this study we showed that Random Forests can deal with large number of weak input variables without overfitting. It can account for non-additive interactions between these input variables. Random Forests can also be used for variable selection without being adversely affected by collinearities. ^ Random Forests can deal with the large-scale data sets without rigorous data preprocessing. It has robust variable importance ranking measure. Proposed is a novel variable selection method in context of Random Forests that uses the data noise level as the cut-off value to determine the subset of the important predictors. This new approach enhanced the ability of the Random Forests algorithm to automatically identify important predictors for complex data. The cut-off value can also be adjusted based on the results of the synthetic positive control experiments. ^ When the data set had high variables to observations ratio, Random Forests complemented the established logistic regression. This study suggested that Random Forests is recommended for such high dimensionality data. One can use Random Forests to select the important variables and then use logistic regression or Random Forests itself to estimate the effect size of the predictors and to classify new observations. ^ We also found that the mean decrease of accuracy is a more reliable variable ranking measurement than mean decrease of Gini. ^

Evaluation of the Hosmer -Lemeshow global goodness -of-fit statistic for mixed variable logistic regression models

The performance of the Hosmer-Lemeshow global goodness-of-fit statistic for logistic regression models was explored in a wide variety of conditions not previously fully investigated. Computer simulations, each consisting of 500 regression models, were run to assess the statistic in 23 different situations. The items which varied among the situations included the number of observations used in each regression, the number of covariates, the degree of dependence among the covariates, the combinations of continuous and discrete variables, and the generation of the values of the dependent variable for model fit or lack of fit.^ The study found that the $\rm\ C$g* statistic was adequate in tests of significance for most situations. However, when testing data which deviate from a logistic model, the statistic has low power to detect such deviation. Although grouping of the estimated probabilities into quantiles from 8 to 30 was studied, the deciles of risk approach was generally sufficient. Subdividing the estimated probabilities into more than 10 quantiles when there are many covariates in the model is not necessary, despite theoretical reasons which suggest otherwise. Because it does not follow a X$\sp2$ distribution, the statistic is not recommended for use in models containing only categorical variables with a limited number of covariate patterns.^ The statistic performed adequately when there were at least 10 observations per quantile. Large numbers of observations per quantile did not lead to incorrect conclusions that the model did not fit the data when it actually did. However, the statistic failed to detect lack of fit when it existed and should be supplemented with further tests for the influence of individual observations. Careful examination of the parameter estimates is also essential since the statistic did not perform as desired when there was moderate to severe collinearity among covariates.^ Two methods studied for handling tied values of the estimated probabilities made only a slight difference in conclusions about model fit. Neither method split observations with identical probabilities into different quantiles. Approaches which create equal size groups by separating ties should be avoided. ^

Activity of anidulafungin in a murine model of Candida krusei infection: evaluation of mortality and disease burden by quantitative tissue cultures and measurement of serum (1,3)-beta-D-glucan levels.

Experience with anidulafungin against Candida krusei is limited. Immunosuppressed mice were injected with 1.3 x 10(7) to 1.5 x 10(7) CFU of C. krusei. Animals were treated with saline, 40 mg/kg fluconazole, 1 mg/kg amphotericin B, or 10 and 20 mg/kg anidulafungin for 5 days. Anidulafungin improved survival and significantly reduced the number of CFU/g in kidneys and serum beta-glucan levels.

Design and evaluation of the parent education component of a comprehensive violence prevention project in middle schools

Low parental monitoring is related to youth risk behaviors such as delinquency and aggression. The purpose of this dissertation was to describe the development and evaluation of a parent education intervention to increase parental monitoring in Hispanic parents of middle school children.^ The first study described the process of intervention mapping as used to develop Padres Trabajando por la Paz, a newsletter intervention for parents. Using theory, empirical literature, and information from the target population, performance objectives and determinants for monitoring were defined. Learning objectives were specified and a staged social-cognitive approach was used to develop methods and strategies delivered through newsletters.^ The second study examined the outcomes of a randomized trial of the newsletter intervention. Outcome measures consisted of a general measure of monitoring, parent and child reports of monitoring behaviors targeted by the intervention, and psychosocial determinants of monitoring (self-efficacy, norms, outcome expectancies, knowledge, and beliefs). Seventy-seven parents completed the randomized trial, half of which received four newsletters over an eight-week period. Results revealed a significant interaction effect for baseline and treatment for parent's reports of norms for monitoring (p =.009). Parents in the experimental condition who scored low at baseline reported increased norms for monitoring at follow-up. A significant interaction effect for child reports of parental monitoring behaviors (p =.04) reflected an small increase across baseline levels in the experimental condition and decreases for the control condition at higher baseline scores. Both groups of parents reported increased levels of monitoring at follow-up. No other outcome measures varied significantly by condition.^ The third study examined the relationship between the psychosocial determinants of parental monitoring and parental monitoring behaviors in the study population. Weak evidence for a relationship between outcome expectancies and parental monitoring behaviors suggests further research in the area utilizing stronger empirical models such as longitudinal design and structural equation modeling.^ The low-cost, minimal newsletter intervention showed promise for changing norms among Hispanic parents for parental monitoring. In light of the importance of parental monitoring as a protective factor for youth health risk behaviors, more research needs to be done to develop and evaluate interventions to increase parental monitoring. ^

A MANAGEMENT INFORMATION SYSTEM AND EVALUATION SCHEME FOR THE UNIVERSITY OF TEXAS EMPLOYEE ASSISTANCE PROGRAMS

Early Employee Assistance Programs (EAPs) had their origin in humanitarian motives, and there was little concern for their cost/benefit ratios; however, as some programs began accumulating data and analyzing it over time, even with single variables such as absenteeism, it became apparent that the humanitarian reasons for a program could be reinforced by cost savings particularly when the existence of the program was subject to justification.^ Today there is general agreement that cost/benefit analyses of EAPs are desirable, but the specific models for such analyses, particularly those making use of sophisticated but simple computer based data management systems, are few.^ The purpose of this research and development project was to develop a method, a design, and a prototype for gathering managing and presenting information about EAPS. This scheme provides information retrieval and analyses relevant to such aspects of EAP operations as: (1) EAP personnel activities, (2) Supervisory training effectiveness, (3) Client population demographics, (4) Assessment and Referral Effectiveness, (5) Treatment network efficacy, (6) Economic worth of the EAP.^ This scheme has been implemented and made operational at The University of Texas Employee Assistance Programs for more than three years.^ Application of the scheme in the various programs has defined certain variables which remained necessary in all programs. Depending on the degree of aggressiveness for data acquisition maintained by program personnel, other program specific variables are also defined. ^

Evaluation of candidate genes for susceptibility to spina bifida

Neural tube defects (NTDs) are the most common severely disabling birth defects in the United States, with a frequency of approximately 1–2 of every 1,000 births. This text includes the identification and evaluation of candidate susceptibility genes that confer risk for the development of neural tube defects (NTDs). The project focused on isolated meningomyelocele, also termed spina bifida (SB). ^ Spina bifida is a complex disease with multifactorial inheritance, therefore the subject population (consisting of North American Caucasians and Hispanics of Mexicali-American descent) was composed of 459 simplex SB families who were tested for genetic associations utilizing the transmission disequilibrium test (TDT), a nonparametric linkage technique. Three categories of candidate genes were studied, including (1) human equivalents of genes determined in mouse models to cause NTDs, (2) HOX and PAX genes, and (3) the MTHFR gene involved in the metabolic pathway of folate. ^ The C677T variant of the 5,10-methylenetetrahydrofolate reductase (MTHFR) gene was the first mutation in this gene to be implicated as a risk factor for NTDs. Our evaluation of the MTHFR gene provides evidence that maternal C677T homozygosity is a risk factor for upper level spina bifida defects in Hispanics [OR = 2.3, P = 0.02]. This observed risk factor is of great importance due to the high prevalence of this homozygous genotype in the Hispanic population. Additionally, maternal C677T/A1298C compound heterozygosity is a risk factor for upper level spina bifida defects in non-Hispanic whites [OR = 3.6, P = 0.03]. ^ For TDT analysis, our total population of 1128 subjects were genotyped for 54 markers from within and/or flanking the 20 candidate genes/gene regions of interest. Significant TDT findings were obtained for 3 of the 54 analyzed markers: d20s101 flanking the PAX1 gene (P = 0.019), d1s228 within the PAX7 gene (P = 0.011), and d2s110 within the PAX8 gene (P = 0.013). These results were followed-up by testing the genes directly for mutations utilizing single-strand conformational analysis (SSCA) and direct sequencing. Multiple variations were detected in each of these PAX genes; however, these variations were not passed from parent to child in phase with the positively transmitted alleles. Therefore, these variations do not contribute to the susceptibility of spina bifida, but rather are previously unreported single nucleotide polymorphisms. ^

The design and evaluation of a 2D dose verification system for intensity modulated radiotherapy

The usage of intensity modulated radiotherapy (IMRT) treatments necessitates a significant amount of patient-specific quality assurance (QA). This research has investigated the precision and accuracy of Kodak EDR2 film measurements for IMRT verifications, the use of comparisons between 2D dose calculations and measurements to improve treatment plan beam models, and the dosimetric impact of delivery errors. New measurement techniques and software were developed and used clinically at M. D. Anderson Cancer Center. The software implemented two new dose comparison parameters, the 2D normalized agreement test (NAT) and the scalar NAT index. A single-film calibration technique using multileaf collimator (MLC) delivery was developed. EDR2 film's optical density response was found to be sensitive to several factors: radiation time, length of time between exposure and processing, and phantom material. Precision of EDR2 film measurements was found to be better than 1%. For IMRT verification, EDR2 film measurements agreed with ion chamber results to 2%/2mm accuracy for single-beam fluence map verifications and to 5%/2mm for transverse plane measurements of complete plan dose distributions. The same system was used to quantitatively optimize the radiation field offset and MLC transmission beam modeling parameters for Varian MLCs. While scalar dose comparison metrics can work well for optimization purposes, the influence of external parameters on the dose discrepancies must be minimized. The ability of 2D verifications to detect delivery errors was tested with simulated data. The dosimetric characteristics of delivery errors were compared to patient-specific clinical IMRT verifications. For the clinical verifications, the NAT index and percent of pixels failing the gamma index were exponentially distributed and dependent upon the measurement phantom but not the treatment site. Delivery errors affecting all beams in the treatment plan were flagged by the NAT index, although delivery errors impacting only one beam could not be differentiated from routine clinical verification discrepancies. Clinical use of this system will flag outliers, allow physicists to examine their causes, and perhaps improve the level of agreement between radiation dose distribution measurements and calculations. The principles used to design and evaluate this system are extensible to future multidimensional dose measurements and comparisons. ^

Evaluation of therapist and client language in Motivational Interviewing (MI) sessions: A secondary analysis of data from the Southern Methodist Alcohol Research Trial (SMART) study

In the United States, “binge” drinking among college students is an emerging public health concern due to the significant physical and psychological effects on young adults. The focus is on identifying interventions that can help decrease high-risk drinking behavior among this group of drinkers. One such intervention is Motivational interviewing (MI), a client-centered therapy that aims at resolving client ambivalence by developing discrepancy and engaging the client in change talk. Of late, there is a growing interest in determining the active ingredients that influence the alliance between the therapist and the client. This study is a secondary analysis of the data obtained from the Southern Methodist Alcohol Research Trial (SMART) project, a dismantling trial of MI and feedback among heavy drinking college students. The present project examines the relationship between therapist and client language in MI sessions on a sample of “binge” drinking college students. Of the 126 SMART tapes, 30 tapes (‘MI with feedback’ group = 15, ‘MI only’ group = 15) were randomly selected for this study. MISC 2.1, a mutually exclusive and exhaustive coding system, was used to code the audio/videotaped MI sessions. Therapist and client language were analyzed for communication characteristics. Overall, therapists adopted a MI consistent style and clients were found to engage in change talk. Counselor acceptance, empathy, spirit, and complex reflections were all significantly related to client change talk (p-values ranged from 0.001 to 0.047). Additionally, therapist ‘advice without permission’ and MI Inconsistent therapist behaviors were strongly correlated with client sustain talk (p-values ranged from 0.006 to 0.048). Simple linear regression models showed a significant correlation between MI consistent (MICO) therapist language (independent variable) and change talk (dependent variable) and MI inconsistent (MIIN) therapist language (independent variable) and sustain talk (dependent variable). The study has several limitations such as small sample size, self-selection bias, poor inter-rater reliability for the global scales and the lack of a temporal measure of therapist and client language. Future studies might consider a larger sample size to obtain more statistical power. In addition the correlation between therapist language, client language and drinking outcome needs to be explored.^

Process evaluation of the Texas occupational safety & health surveillance system

Introduction: The Texas Occupational Safety & Health Surveillance System (TOSHSS) was created to collect, analyze and interpret occupational injury and illness data in order to decrease the impact of occupational injuries within the state of Texas. This process evaluation was performed midway through the 4-year grant to assess the efficiency and effectiveness of the surveillance system’s planning and implementation activities1. ^ Methods: Two evaluation guidelines published by the Centers for Disease Control and Prevention (CDC) were used as the theoretical models for this process evaluation. The Framework for Program Evaluation in Public Health was used to examine the planning and design of TOSHSS using logic models. The Framework for Evaluating Public Health Surveillance Systems was used to examine the implementation of approximately 60 surveillance activities, including uses of the data obtained from the surveillance system. ^ Results/Discussion: TOSHSS planning activities omitted the creation of a scientific advisory committee and specific activities designed to maintain contacts with stakeholders; and proposed activities should be reassessed and aligned with ongoing performance measurement criteria, including the role of collaborators in helping the surveillance system achieve each proposed activity. TOSHSS implementation activities are substantially meeting expectations and received an overall score of 61% for all activities being performed. TOSHSS is considered a surveillance system that is simple, flexible, acceptable, fairly stable, timely, moderately useful, with good data quality and a PVP of 86%. ^ Conclusions: Through the third year of TOSHSS implementation, the surveillance system is has made a considerable contribution to the collection of occupational injury and illness information within the state of Texas. Implementation of the nine recommendations provided under this process evaluation is expected to increase the overall usefulness of the surveillance system and assist TDSHS in reducing occupational fatalities, injuries, and diseases within the state of Texas. ^ 1 Disclaimer: The Texas Occupational Safety and Health Surveillance System is supported by Grant/Cooperative Agreement Number (U60 OH008473-01A1). The content of the current evaluation are solely the responsibility of the authors and do not necessarily represent the official views of the Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health.^