Multilevel Analysis in Household Survey: An Application to Health Condition Data

The aim of this thesis is to apply multilevel regression model in context of household surveys. Hierarchical structure in this type of data is characterized by many small groups. In last years comparative and multilevel analysis in the field of perceived health have grown in size. The purpose of this thesis is to develop a multilevel analysis with three level of hierarchy for Physical Component Summary outcome to: evaluate magnitude of within and between variance at each level (individual, household and municipality); explore which covariates affect on perceived physical health at each level; compare model-based and design-based approach in order to establish informativeness of sampling design; estimate a quantile regression for hierarchical data. The target population are the Italian residents aged 18 years and older. Our study shows a high degree of homogeneity within level 1 units belonging from the same group, with an intraclass correlation of 27% in a level-2 null model. Almost all variance is explained by level 1 covariates. In fact, in our model the explanatory variables having more impact on the outcome are disability, unable to work, age and chronic diseases (18 pathologies). An additional analysis are performed by using novel procedure of analysis :"Linear Quantile Mixed Model", named "Multilevel Linear Quantile Regression", estimate. This give us the possibility to describe more generally the conditional distribution of the response through the estimation of its quantiles, while accounting for the dependence among the observations. This has represented a great advantage of our models with respect to classic multilevel regression. The median regression with random effects reveals to be more efficient than the mean regression in representation of the outcome central tendency. A more detailed analysis of the conditional distribution of the response on other quantiles highlighted a differential effect of some covariate along the distribution.

Comparison of two scoring systems for evaluation of treatment outcome in patients with complete bilateral cleft lip and palate

Objective : To compare two scoring systems: the Huddart/Bodenham system (HB system) and the Bauru-BCLP yardstick (BCLP yardstick), which classify treatment outcome in terms of dental arch relationships in patients with complete bilateral cleft lip and palate (CBCLP). The predictive value of these scoring systems for treatment outcome was also evaluated. Design : Retrospective longitudinal study. Patients : Dental arch relationships of 43 CBCLP patients were evaluated at 6, 9, and 12 years. Setting : Treatment outcome in BCLP patients using two scoring systems. Main Outcome Measures : For each age group, the HB scores were correlated with the BCLP yardstick scores using Spearman's correlation coefficient. The predictive value of the two scoring systems was evaluated by backward regression analysis. Results : Intraobserver Kappa values for the BCLP yardstick scoring for the two observers were .506 and .627, respectively, and the interobserver reliability ranged from .427 and .581. The intraobserver reliability for the HB system ranged from .92 to .97 and the interobserver reliability from .88 to .96. The BCLP yardstick scores of 6 and 9 years together were predictors for the outcome at 12 years (explained variance 41.3%). Adding the incisor and lateral HB scores in the regression model increased the explained variance to 67%. Conclusions : The BCLP yardstick and the HB system are reliable scoring systems for evaluation of dental arch relationships of CBCLP patients. The HB system categorizes treatment outcome into similar categories as the BCLP yardstick. In case a more sensitive measure of treatment outcome is needed, selectively both scoring systems should be used.

A prediction model for assessing residential radon concentration in Switzerland

Indoor radon is regularly measured in Switzerland. However, a nationwide model to predict residential radon levels has not been developed. The aim of this study was to develop a prediction model to assess indoor radon concentrations in Switzerland. The model was based on 44,631 measurements from the nationwide Swiss radon database collected between 1994 and 2004. Of these, 80% randomly selected measurements were used for model development and the remaining 20% for an independent model validation. A multivariable log-linear regression model was fitted and relevant predictors selected according to evidence from the literature, the adjusted R², the Akaike's information criterion (AIC), and the Bayesian information criterion (BIC). The prediction model was evaluated by calculating Spearman rank correlation between measured and predicted values. Additionally, the predicted values were categorised into three categories (50th, 50th-90th and 90th percentile) and compared with measured categories using a weighted Kappa statistic. The most relevant predictors for indoor radon levels were tectonic units and year of construction of the building, followed by soil texture, degree of urbanisation, floor of the building where the measurement was taken and housing type (P-values <0.001 for all). Mean predicted radon values (geometric mean) were 66 Bq/m³ (interquartile range 40-111 Bq/m³) in the lowest exposure category, 126 Bq/m³ (69-215 Bq/m³) in the medium category, and 219 Bq/m³ (108-427 Bq/m³) in the highest category. Spearman correlation between predictions and measurements was 0.45 (95%-CI: 0.44; 0.46) for the development dataset and 0.44 (95%-CI: 0.42; 0.46) for the validation dataset. Kappa coefficients were 0.31 for the development and 0.30 for the validation dataset, respectively. The model explained 20% overall variability (adjusted R²). In conclusion, this residential radon prediction model, based on a large number of measurements, was demonstrated to be robust through validation with an independent dataset. The model is appropriate for predicting radon level exposure of the Swiss population in epidemiological research. Nevertheless, some exposure misclassification and regression to the mean is unavoidable and should be taken into account in future applications of the model.

Mother-Child Spiritual Discourse: A Mixed Methods Study

Parents and children, starting at very young ages, discuss religious and spiritual issues¿where we come from, what happens to us after we die, is there a God, and so on. Unfortunately, few studies have analyzed the content and structure of parent-child conversation about religion and spirituality (Boyatzis & Janicki, 2003; Dollahite & Thatcher, 2009), and most studies have relied on self-report with no direct observation. The current study examined mother-child (M-C) spiritual discourse to learn about its content, structure, and frequency through a survey inventory in combination with direct video observation using a novel structured task. We also analyzed how mothers¿ religiosity along several major dimensions related to their communication behaviors within both methods. Mothers (N = 39, M age = 40) of children aged 3-12 completed a survey packet on M-C spiritual discourse and standard measures of mothers¿ religious fundamentalism, intrinsic religiosity, sanctification of parenting (how much the mother saw herself as doing God¿s work as a parent), and a new measure of parental openness to children¿s spirituality. Then, in a structured task in our lab, mothers (N = 33) and children (M age = 7.33) watched a short film or read a short book that explored death in an age-appropriate manner and then engaged in a videotaped conversation about the movie or book and their religious or spiritual beliefs. Frequency of M-C spiritual discourse was positively related to mothers¿ religious fundamentalism (r = .71, p = .00), intrinsic religiosity (r = .77, p = .00), and sanctification of parenting (r = .79, p = .00), but, surprisingly, was inversely related to mothers¿ v openness to child¿s spirituality (r = -.52, p = .00). Survey data showed that the two most common topics discussed were God (once a week) and religion as it relates to moral issues (once a week). According to mothers their children¿s most common method of initiating spiritual discourse was to repeat what he or she has heard parents or family say about religious issues (M = 2.97; once a week); mothers¿ most common method was to describe their own religious/spiritual beliefs (M = 2.92). Spiritual discourse most commonly occurred either at bedtime or mealtime as reported by 26% of mothers, with the most common triggers reported as daily routine/random thoughts (once a week) and observations of nature (once a week). Mothers¿ most important goals for spiritual discourse were to let their children know that they love them (M = 3.72; very important) and to help them become a good and moral person (M = 3.67; very important). A regression model showed that significant variance in frequency of mother-child spiritual discourse (R2 = .84, p = .00) was predicted by the mothers¿ importance of goals during discourse (ß = 0.46, p = .00), frequency that the mother¿s spirituality was deepened through spiritual discourse (ß = 0.39, p = .00), and the mother¿s fundamentalism (ß = 0.20, p = .05). In a separate regression, the mother¿s comfort in the structured task (ß = 0.70, p = .00), and the number of open-ended questions she asked (ß = -0.26, p = .03) predicted the reciprocity between mother and child (R2 = .62, p = .00). In addition, the mother¿s age (ß = 0.22, p = .059) and comfort during the task (ß = 0.73, p = .00) predicted the child¿s engagement within the structured task. Other findings and theoretical and methodological implications will be discussed.

Model Evaluation Based on the Distribution of Estimated Absolute Prediction Error

The construction of a reliable, practically useful prediction rule for future response is heavily dependent on the "adequacy" of the fitted regression model. In this article, we consider the absolute prediction error, the expected value of the absolute difference between the future and predicted responses, as the model evaluation criterion. This prediction error is easier to interpret than the average squared error and is equivalent to the mis-classification error for the binary outcome. We show that the distributions of the apparent error and its cross-validation counterparts are approximately normal even under a misspecified fitted model. When the prediction rule is "unsmooth", the variance of the above normal distribution can be estimated well via a perturbation-resampling method. We also show how to approximate the distribution of the difference of the estimated prediction errors from two competing models. With two real examples, we demonstrate that the resulting interval estimates for prediction errors provide much more information about model adequacy than the point estimates alone.

Additive Hazards Models with Latent Treatment Effectiveness Lag Time

In many clinical trials to evaluate treatment efficacy, it is believed that there may exist latent treatment effectiveness lag times after which medical procedure or chemical compound would be in full effect. In this article, semiparametric regression models are proposed and studied to estimate the treatment effect accounting for such latent lag times. The new models take advantage of the invariance property of the additive hazards model in marginalizing over random effects, so parameters in the models are easy to be estimated and interpreted, while the flexibility without specifying baseline hazard function is kept. Monte Carlo simulation studies demonstrate the appropriateness of the proposed semiparametric estimation procedure. Data collected in the actual randomized clinical trial, which evaluates the effectiveness of biodegradable carmustine polymers for treatment of recurrent brain tumors, are analyzed.

Checking Assumptions in Latent Class Regression Models via a Markov Chain Monte Carlo Estimation Approach: An Application to Depression and Socio-Economic Status

Latent class regression models are useful tools for assessing associations between covariates and latent variables. However, evaluation of key model assumptions cannot be performed using methods from standard regression models due to the unobserved nature of latent outcome variables. This paper presents graphical diagnostic tools to evaluate whether or not latent class regression models adhere to standard assumptions of the model: conditional independence and non-differential measurement. An integral part of these methods is the use of a Markov Chain Monte Carlo estimation procedure. Unlike standard maximum likelihood implementations for latent class regression model estimation, the MCMC approach allows us to calculate posterior distributions and point estimates of any functions of parameters. It is this convenience that allows us to provide the diagnostic methods that we introduce. As a motivating example we present an analysis focusing on the association between depression and socioeconomic status, using data from the Epidemiologic Catchment Area study. We consider a latent class regression analysis investigating the association between depression and socioeconomic status measures, where the latent variable depression is regressed on education and income indicators, in addition to age, gender, and marital status variables. While the fitted latent class regression model yields interesting results, the model parameters are found to be invalid due to the violation of model assumptions. The violation of these assumptions is clearly identified by the presented diagnostic plots. These methods can be applied to standard latent class and latent class regression models, and the general principle can be extended to evaluate model assumptions in other types of models.

Bayesian Hierarchical Distributed Lag Models for Summer Ozone Exposure and Cardio-Respiratory Mortality

In this paper, we develop Bayesian hierarchical distributed lag models for estimating associations between daily variations in summer ozone levels and daily variations in cardiovascular and respiratory (CVDRESP) mortality counts for 19 U.S. large cities included in the National Morbidity Mortality Air Pollution Study (NMMAPS) for the period 1987 - 1994. At the first stage, we define a semi-parametric distributed lag Poisson regression model to estimate city-specific relative rates of CVDRESP associated with short-term exposure to summer ozone. At the second stage, we specify a class of distributions for the true city-specific relative rates to estimate an overall effect by taking into account the variability within and across cities. We perform the calculations with respect to several random effects distributions (normal, t-student, and mixture of normal), thus relaxing the common assumption of a two-stage normal-normal hierarchical model. We assess the sensitivity of the results to: 1) lag structure for ozone exposure; 2) degree of adjustment for long-term trends; 3) inclusion of other pollutants in the model;4) heat waves; 5) random effects distributions; and 6) prior hyperparameters. On average across cities, we found that a 10ppb increase in summer ozone level for every day in the previous week is associated with 1.25 percent increase in CVDRESP mortality (95% posterior regions: 0.47, 2.03). The relative rate estimates are also positive and statistically significant at lags 0, 1, and 2. We found that associations between summer ozone and CVDRESP mortality are sensitive to the confounding adjustment for PM_10, but are robust to: 1) the adjustment for long-term trends, other gaseous pollutants (NO_2, SO_2, and CO); 2) the distributional assumptions at the second stage of the hierarchical model; and 3) the prior distributions on all unknown parameters. Bayesian hierarchical distributed lag models and their application to the NMMAPS data allow us estimation of an acute health effect associated with exposure to ambient air pollution in the last few days on average across several locations. The application of these methods and the systematic assessment of the sensitivity of findings to model assumptions provide important epidemiological evidence for future air quality regulations.

Modeling multilevel sleep transitional data via Poisson log-linear multilevel models

This paper proposes Poisson log-linear multilevel models to investigate population variability in sleep state transition rates. We specifically propose a Bayesian Poisson regression model that is more flexible, scalable to larger studies, and easily fit than other attempts in the literature. We further use hierarchical random effects to account for pairings of individuals and repeated measures within those individuals, as comparing diseased to non-diseased subjects while minimizing bias is of epidemiologic importance. We estimate essentially non-parametric piecewise constant hazards and smooth them, and allow for time varying covariates and segment of the night comparisons. The Bayesian Poisson regression is justified through a re-derivation of a classical algebraic likelihood equivalence of Poisson regression with a log(time) offset and survival regression assuming piecewise constant hazards. This relationship allows us to synthesize two methods currently used to analyze sleep transition phenomena: stratified multi-state proportional hazards models and log-linear models with GEE for transition counts. An example data set from the Sleep Heart Health Study is analyzed.

Towards greater accuracy in individual-tree mortality regression

Background mortality is an essential component of any forest growth and yield model. Forecasts of mortality contribute largely to the variability and accuracy of model predictions at the tree, stand and forest level. In the present study, I implement and evaluate state-of-the-art techniques to increase the accuracy of individual tree mortality models, similar to those used in many of the current variants of the Forest Vegetation Simulator, using data from North Idaho and Montana. The first technique addresses methods to correct for bias induced by measurement error typically present in competition variables. The second implements survival regression and evaluates its performance against the traditional logistic regression approach. I selected the regression calibration (RC) algorithm as a good candidate for addressing the measurement error problem. Two logistic regression models for each species were fitted, one ignoring the measurement error, which is the “naïve” approach, and the other applying RC. The models fitted with RC outperformed the naïve models in terms of discrimination when the competition variable was found to be statistically significant. The effect of RC was more obvious where measurement error variance was large and for more shade-intolerant species. The process of model fitting and variable selection revealed that past emphasis on DBH as a predictor variable for mortality, while producing models with strong metrics of fit, may make models less generalizable. The evaluation of the error variance estimator developed by Stage and Wykoff (1998), and core to the implementation of RC, in different spatial patterns and diameter distributions, revealed that the Stage and Wykoff estimate notably overestimated the true variance in all simulated stands, but those that are clustered. Results show a systematic bias even when all the assumptions made by the authors are guaranteed. I argue that this is the result of the Poisson-based estimate ignoring the overlapping area of potential plots around a tree. Effects, especially in the application phase, of the variance estimate justify suggested future efforts of improving the accuracy of the variance estimate. The second technique implemented and evaluated is a survival regression model that accounts for the time dependent nature of variables, such as diameter and competition variables, and the interval-censored nature of data collected from remeasured plots. The performance of the model is compared with the traditional logistic regression model as a tool to predict individual tree mortality. Validation of both approaches shows that the survival regression approach discriminates better between dead and alive trees for all species. In conclusion, I showed that the proposed techniques do increase the accuracy of individual tree mortality models, and are a promising first step towards the next generation of background mortality models. I have also identified the next steps to undertake in order to advance mortality models further.

Adjudication-related processes are underreported and lack standardization in clinical trials of venous thromboembolism: a systematic review

OBJECTIVES Although the use of an adjudication committee (AC) for outcomes is recommended in randomized controlled trials, there are limited data on the process of adjudication. We therefore aimed to assess whether the reporting of the adjudication process in venous thromboembolism (VTE) trials meets existing quality standards and which characteristics of trials influence the use of an AC. STUDY DESIGN AND SETTING We systematically searched MEDLINE and the Cochrane Library from January 1, 2003, to June 1, 2012, for randomized controlled trials on VTE. We abstracted information about characteristics and quality of trials and reporting of adjudication processes. We used stepwise backward logistic regression model to identify trial characteristics independently associated with the use of an AC. RESULTS We included 161 trials. Of these, 68.9% (111 of 161) reported the use of an AC. Overall, 99.1% (110 of 111) of trials with an AC used independent or blinded ACs, 14.4% (16 of 111) reported how the adjudication decision was reached within the AC, and 4.5% (5 of 111) reported on whether the reliability of adjudication was assessed. In multivariate analyses, multicenter trials [odds ratio (OR), 8.6; 95% confidence interval (CI): 2.7, 27.8], use of a data safety-monitoring board (OR, 3.7; 95% CI: 1.2, 11.6), and VTE as the primary outcome (OR, 5.7; 95% CI: 1.7, 19.4) were associated with the use of an AC. Trials without random allocation concealment (OR, 0.3; 95% CI: 0.1, 0.8) and open-label trials (OR, 0.3; 95% CI: 0.1, 1.0) were less likely to report an AC. CONCLUSION Recommended processes of adjudication are underreported and lack standardization in VTE-related clinical trials. The use of an AC varies substantially by trial characteristics.

IN-VIVO CT DOSIMETRY DURING VIRTUAL COLONOSCOPY

Virtual colonoscopy (VC) is a minimally invasive means for identifying colorectal polyps and colorectal lesions by insufflating a patient’s bowel, applying contrast agent via rectal catheter, and performing multi-detector computed tomography (MDCT) scans. The technique is recommended for colonic health screening by the American Cancer Society but not funded by the Centers for Medicare and Medicaid Services (CMS) partially because of potential risks from radiation exposure. To date, no in‐vivo organ dose measurements have been performed for MDCT scans; thus, the accuracy of any current dose estimates is currently unknown. In this study, two TLDs were affixed to the inner lumen of standard rectal catheters used in VC, and in-vivo rectal dose measurements were obtained within 6 VC patients. In order to calculate rectal dose, TLD-100 powder response was characterized at diagnostic doses such that appropriate correction factors could be determined for VC. A third-order polynomial regression with a goodness of fit factor of R2=0.992 was constructed from this data. Rectal dose measurements were acquired with TLDs during simulated VC within a modified anthropomorphic phantom configured to represent three sizes of patients undergoing VC. The measured rectal doses decreased in an exponential manner with increasing phantom effective diameter, with R2=0.993 for the exponential regression model and a maximum percent coefficient of variation (%CoV) of 4.33%. In-vivo measurements yielded rectal doses ranged from that decreased exponentially with increasing patient effective diameter, in a manner that was also favorably predicted by the size specific dose estimate (SSDE) model for all VC patients that were of similar age, body composition, and TLD placement. The measured rectal dose within a younger patient was favorably predicted by the anthropomorphic phantom dose regression model due to similarities in the percentages of highly attenuating material at the respective measurement locations and in the placement of the TLDs. The in-vivo TLD response did not increase in %CoV with decreasing dose, and the largest %CoV was 10.0%.

Hypofibrinolysis in type 2 diabetes: the role of the inflammatory pathway and complement C3

AIMS/HYPOTHESIS Plasminogen activator inhibitor-1 (PAI-1) has been regarded as the main antifibrinolytic protein in diabetes, but recent work indicates that complement C3 (C3), an inflammatory protein, directly compromises fibrinolysis in type 1 diabetes. The aim of the current project was to investigate associations between C3 and fibrinolysis in a large cohort of individuals with type 2 diabetes. METHODS Plasma levels of C3, C-reactive protein (CRP), PAI-1 and fibrinogen were analysed by ELISA in 837 patients enrolled in the Edinburgh Type 2 Diabetes Study. Fibrin clot lysis was analysed using a validated turbidimetric assay. RESULTS Clot lysis time correlated with C3 and PAI-1 plasma levels (r = 0.24, p < 0.001 and r = 0.22, p < 0.001, respectively). In a multivariable regression model involving age, sex, BMI, C3, PAI-1, CRP and fibrinogen, and using log-transformed data as appropriate, C3 was associated with clot lysis time (regression coefficient 0.227 [95% CI 0.161, 0.292], p < 0.001), as was PAI-1 (regression coefficient 0.033 [95% CI 0.020, 0.064], p < 0.05) but not fibrinogen (regression coefficient 0.003 [95% CI -0.046, 0.051], p = 0.92) or CRP (regression coefficient 0.024 [95% CI -0.008, 0.056], p = 0.14). No correlation was demonstrated between plasma levels of C3 and PAI-1 (r = -0.03, p = 0.44), consistent with previous observations that the two proteins affect different pathways in the fibrinolytic system. CONCLUSIONS/INTERPRETATION Similarly to PAI-1, C3 plasma levels are independently associated with fibrin clot lysis in individuals with type 2 diabetes. Therefore, future studies should analyse C3 plasma levels as a surrogate marker of fibrinolysis potential in this population.

Efficiency Bounds for Estimating Linear Functionals of Nonparametric Regression Models with Endogenous Regressors

Consider a nonparametric regression model Y=mu*(X) + e, where the explanatory variables X are endogenous and e satisfies the conditional moment restriction E[e|W]=0 w.p.1 for instrumental variables W. It is well known that in these models the structural parameter mu* is 'ill-posed' in the sense that the function mapping the data to mu* is not continuous. In this paper, we derive the efficiency bounds for estimating linear functionals E[p(X)mu*(X)] and int_{supp(X)}p(x)mu*(x)dx, where p is a known weight function and supp(X) the support of X, without assuming mu* to be well-posed or even identified.

Relationship between dietary patterns and body mass index in Hispanic children ages 4--19 years of age

Background. In over 30 years, the prevalence of overweight for children and adolescents has increased across the United States (Barlow et al., 2007; Ogden, Flegal, Carroll, & Johnson, 2002). Childhood obesity is linked with adverse physiological and psychological issues in youth and affects ethnic/minority populations in disproportionate rates (Barlow et al., 2007; Butte et al., 2006; Butte, Cai, Cole, Wilson, Fisher, Zakeri, Ellis, & Comuzzie, 2007). More importantly, overweight in children and youth tends to track into adulthood (McNaughton, Ball, Mishra, & Crawford, 2008; Ogden et al., 2002). Childhood obesity affects body functions such as the cardiovascular, respiratory, gastrointestinal, and endocrine systems, including emotional health (Barlow et al., 2007, Ogden et al., 2002). Several dietary factors have been associated with the development of obesity in children; however, these factors have not been fully elucidated, especially in ethnic/minority children. In particular, few studies have been done to determine the effects of different meal patterns on the development of obesity in children. Purpose. The purpose of the study is to examine the relationships between daily proportions of energy consumed and energy derived from fat across breakfast, lunch, dinner, and snack, and obesity among Hispanic children and adolescents. Methods. A cross-sectional design was used to evaluate the relationship between dietary patterns and overweight status in Hispanic children and adolescents 4-19 years of age who participated in the Viva La Familia Study. The goal of the Viva La Familia Study was to evaluate genetic and environmental factors affecting childhood obesity and its co-morbidities in the Hispanic population (Butte et al., 2006, 2007). The study enrolled 1030 Hispanic children and adolescents from 319 families and examined factors related to increased body weight by focusing on a multilevel analysis of extensive sociodemographic, genetic, metabolic, and behavioral data. Baseline dietary intakes of the children were collected using 24-hour recalls, and body mass index was calculated from measured height and weight, and classified using the CDC standards. Dietary data were analyzed using a GEE population-averaged panel-data model with a cluster variable family identifier to include possible correlations within related data sets. A linear regression model was used to analyze associations of dietary patterns using possible covariates, and to examine the percentage of daily energy coming from breakfast, lunch, dinner, and snack while adjusting for age, sex, and BMI z-score. Random-effects logistic regression models were used to determine the relationship of the dietary variables with obesity status and to understand if the percent energy intake (%EI) derived from fat from all meals (breakfast, lunch, dinner, and snacks) affected obesity. Results. Older children (age 4-19 years) consumed a higher percent of energy at lunch and dinner and less percent energy from snacks compared to younger children. Age was significantly associated with percentage of total energy intake (%TEI) for lunch, as well as dinner, while no association was found by gender. Percent of energy consumed from dinner significantly differed by obesity status, with obese children consuming more energy at dinner (p = 0.03), but no associations were found between percent energy from fat and obesity across all meals. Conclusions. Information from this study can be used to develop interventions that target dietary intake patterns in obesity prevention programs for Hispanic children and adolescents. In particular, intervention programs for children should target dietary patterns with energy intake that is spread throughout the day and earlier in the day. These results indicate that a longitudinal study should be used to further explore the relationship of dietary patterns and BMI in this and other populations (Dubois et al., 2008; Rodriquez & Moreno, 2006; Thompson et al., 2005; Wilson et al., in review, 2008). ^