Factors that affect non-adherence to first line antiretroviral medications among HIV infected children and adolescents in Botswana

The aims of the study were to determine the prevalence of and factors that affect non-adherence to first line antiretroviral (ARV) medications among HIV infected children and adolescents in Botswana. The study used secondary data from Botswana-Baylor Children's Clinical Center of Excellence for the period of June 2008 to February 10th, 2010. The study design was cross-sectional and case-comparison between non-adherent and adherent participants was used to examine the effects of socio-demographic and medication factors on non-adherence to ARV medications. A case was defined as non-adherent child with adherence level < 95% based on pill count and measurement of liquid formulations. The comparison group consisted of children with adherence levels ≥95%.^ A total of 842 participants met the eligibility criteria for determination of the prevalence of non-adherence and 338 participants (169 cases and 169 individuals) were used in the analysis to estimate the effects of factors on non-adherence. ^ Univariate and multivariable logistic regression were used to estimate the association between non-adherence (outcome) and socio-demographic and medication factors (exposures). The prevalence of non-adherence for participants on first line ARV medications was 20.0% (169/842).^ Increase in age (OR (95% CI): 1.10 (1.04–1.17) p = 0.001) was associated with nonadherence, while increase in number of caregivers (OR (95% CI): 0.72 (0.56–0.93) p = 0.01) and increase in number of monthly visits (OR (95% CI): 0.92 (0.86–0.99) p = 0.02), were associated with good adherence in both the unadjusted and the adjusted models. For the categorical variables, having more than two caregivers (OR (95% CI): 0.66 (0.28–0.84), p = 0.002) was associated with good adherence even in the adjusted model. ^ Conclusion. The prevalence of non-adherence to antiretroviral medicines among the study population was estimated to be 20.0%. In previous studies, adherence levels of ≥ 95% have been associated with better clinical outcomes and suppression of virus to prevent development of resistance. Older age, fewer numbers of caregivers and fewer monthly visits were associated with non-adherence. Strategies to improve and sustain adherence especially among older children are needed. The role of caregivers and social support should be investigated further.^

Factors associated with needle sharing among Black male injection drug users in Harris County, Texas

Background. Injection drug users (IDUs) are at increased risk for HIV transmission due to unique risk behaviors, such as sharing needles. In Houston, IDUs account for 18% of all HIV/AIDS cases among Black males. ^ Objectives. This analysis compared demographic, behavioral, and psychosocial characteristics of needle sharing and non-sharing IDUs in a population of Black males in Harris County, Texas. ^ Methods. Data used for this analysis were from the second IDU cycle of the National HIV Behavioral Surveillance System. This dataset included a sample of 288 Black male IDUs. Univariate and multivariate statistical analysis were performed to determine statistically significant associations of needle sharing in this population and to create a functional model to inform local HIV prevention programs. ^ Results. Half of the participants in this analysis shared needles in the past 12 months. Compared to non-sharers, sharers were more likely to be homeless (OR=3.70, p<0.01) or arrested in the past year (OR=2.31, p<0.01), inject cocaine (OR=2.07, p<0.01), report male-to-male sex in the past year (OR=6.97, p<0.01), and to exchange sex for money or drugs. Sharers were less likely than non-sharers to graduate high school (OR=0.36, p<0.01), earn $5,000 or more a year (OR=1.15, p=0.05), get needles from a medical source (OR=0.59, p=0.03), and ever test for HIV (OR=0.17, p<0.01). Sharers were more likely to report depressive symptoms (OR=3.49, p<0.01), lower scores on the family support scale (mean difference 0.41, p=0.01) and decision-making confidence scale (mean difference 0.38, p<0.01), and greater risk-taking (mean difference -0.49, p<0.01) than non-sharers. In a multivariable logistic regression, sharers were less likely to have graduated high school (OR=0.33, p<0.01) and have been tested for HIV (OR=0.12, p<0.01) and were more likely to have been arrested in the past year (OR=2.3, p<0.01), get needles from a street source (OR=3.87, p<0.01), report male-to-male sex (OR=7.01, p<0.01), and have depressive symptoms (OR=2.36, p=0.02) and increased risk-taking (OR=1.78, p=0.01). ^ Conclusions. IDUs that shared needles are different from those that did not, reporting lower socioeconomic status, increased sexual and risk behaviors, increased depressive symptoms and increased risk-taking. These findings suggest that intervention programs that also address these demographic, behavioral, and psychosocial factors may be more successful in decreasing needle sharing among this population.^

Genome-wide association study of LDL-cholesterol over ApoB ratio

Genome-Wide Association Study analytical (GWAS) methods were applied in a large biracial sample of individuals to investigate variation across the genome for its association with a surrogate low-density lipoprotein (LDL) particle size phenotype, the ratio of LDL-cholesterol level over ApoB level. Genotyping was performed on the Affymetrix 6.0 GeneChip with approximately one million single nucleotide polymorphisms (SNPs). The ratio of LDL cholesterol to ApoB was calculated, and association tests used multivariable linear regression analysis with an additive genetic model after adjustment for the covariates sex, age and BMI. Association tests were performed separately in African Americans and Caucasians. There were 9,562 qualified individuals in the Caucasian group and 3,015 qualified individuals in the African American group. Overall, in Caucasians two statistically significant loci were identified as being associated with the ratio of LDL-cholesterol over ApoB: rs10488699 (p<5 x10-8, 11q23.3 near BUD13) and the SNP rs964184 (p<5 x10-8 11q23.3 near ZNF259). We also found rs12286037 ((p<4x10-7) (11q23.3) near APOA5/A4/C3/A1 with suggestive associate in the Caucasian sample. In exploratory analyses, a difference in the pattern of association between individuals taking and not taking LDL-cholesterol lowering medications was observed. Individuals who were not taking medications had smaller p-value than those taking medication. In the African-American group, there were no significant (p<5x10-8) or suggestive associations (p<4x10-7) with the ratio of LDL-cholesterol over ApoB after adjusting for age, BMI, and sex and comparing individuals with and without LDL-cholesterol lowering medication. Conclusions: There were significant and suggestive associations between SNP genotype and the ratio of LDL-cholesterol to ApoB in Caucasians, but these associations may be modified by medication treatment.^

Assessment of the effect on statistical power of regression model misspecification by using techniques of mathematical statistics and simulation study

Objectives. This paper seeks to assess the effect on statistical power of regression model misspecification in a variety of situations. ^ Methods and results. The effect of misspecification in regression can be approximated by evaluating the correlation between the correct specification and the misspecification of the outcome variable (Harris 2010).In this paper, three misspecified models (linear, categorical and fractional polynomial) were considered. In the first section, the mathematical method of calculating the correlation between correct and misspecified models with simple mathematical forms was derived and demonstrated. In the second section, data from the National Health and Nutrition Examination Survey (NHANES 2007-2008) were used to examine such correlations. Our study shows that comparing to linear or categorical models, the fractional polynomial models, with the higher correlations, provided a better approximation of the true relationship, which was illustrated by LOESS regression. In the third section, we present the results of simulation studies that demonstrate overall misspecification in regression can produce marked decreases in power with small sample sizes. However, the categorical model had greatest power, ranging from 0.877 to 0.936 depending on sample size and outcome variable used. The power of fractional polynomial model was close to that of linear model, which ranged from 0.69 to 0.83, and appeared to be affected by the increased degrees of freedom of this model.^ Conclusion. Correlations between alternative model specifications can be used to provide a good approximation of the effect on statistical power of misspecification when the sample size is large. When model specifications have known simple mathematical forms, such correlations can be calculated mathematically. Actual public health data from NHANES 2007-2008 were used as examples to demonstrate the situations with unknown or complex correct model specification. Simulation of power for misspecified models confirmed the results based on correlation methods but also illustrated the effect of model degrees of freedom on power.^

Micronutrient intakes and their associations with markers of inflammation, subclinical atherosclerosis, cardiovascular disease, type II diabetes and metabolic syndrome

We investigated cross-sectional associations between intakes of zinc, magnesium, heme- and non heme iron, beta-carotene, vitamin C and vitamin E and inflammation and subclinical atherosclerosis in the Multi-Ethnic Study of Atherosclerosis (MESA). We also investigated prospective associations between those micronutrients and incident MetS, T2D and CVD. Participants between 45-84 years of age at baseline were followed between 2000 and 2007. Dietary intake was assessed at baseline using a 120-item food frequency questionnaire. Multivariable linear regression and Cox proportional hazard regression models were used to evaluate associations of interest. Dietary intakes of non-heme iron and Mg were inversely associated with tHcy concentrations (geometric means across quintiles: 9.11, 8.86, 8.74, 8.71, and 8.50 µmol/L for non-heme iron, and 9.20, 9.00, 8.65, 8.76, and 8.33 µmol/L for Mg; ptrends <0.001). Mg intake was inversely associated with high CC-IMT; odds ratio (95% CI) for extreme quintiles 0.76 (0.58, 1.01), ptrend: 0.002. Dietary Zn and heme-iron were positively associated with CRP (geometric means: 1.73, 1.75, 1.78, 1.88, and 1.96 mg/L for Zn and 1.72, 1.76, 1.83, 1.86, and 1.94 mg/L for heme-iron). In the prospective analysis, dietary vitamin E intake was inversely associated with incident MetS and with incident CVD (HR [CI] for extreme quintiles - MetS: 0.78 [0.62-0.97] ptrend=0.01; CVD: 0.69 [0.46-1.03]; ptrend =0.04). Intake of heme-iron from red meat and Zn from red meat, but not from other sources, were each positively associated with risk of CVD (HR [CI] - heme-iron from red meat: 1.65 [1.10-2.47] ptrend = 0.01; Zn from red meat: 1.51 [1.02 - 2.24] ptrend =0.01) and MetS (HR [CI] - heme-iron from red meat: 1.25 [0.99-1.56] ptrend =0.03; Zn from red meat: 1.29 [1.03-1.61]; ptrend = 0.04). All associations evaluated were similar across different strata of gender, race-ethnicity and alcohol intake. Most of the micronutrients investigated were not associated with the outcomes of interest in this multi-ethnic cohort. These observations do not provide consistent support for the hypothesized association of individual nutrients with inflammatory markers, MetS, T2D, or CVD. However, nutrients consumed in red meat, or consumption of red meat as a whole, may increase risk of MetS and CVD.^

Effects of informative censoring on the hazard function: Application to the proportional hazards regression model

The standard analyses of survival data involve the assumption that survival and censoring are independent. When censoring and survival are related, the phenomenon is known as informative censoring. This paper examines the effects of an informative censoring assumption on the hazard function and the estimated hazard ratio provided by the Cox model.^ The limiting factor in all analyses of informative censoring is the problem of non-identifiability. Non-identifiability implies that it is impossible to distinguish a situation in which censoring and death are independent from one in which there is dependence. However, it is possible that informative censoring occurs. Examination of the literature indicates how others have approached the problem and covers the relevant theoretical background.^ Three models are examined in detail. The first model uses conditionally independent marginal hazards to obtain the unconditional survival function and hazards. The second model is based on the Gumbel Type A method for combining independent marginal distributions into bivariate distributions using a dependency parameter. Finally, a formulation based on a compartmental model is presented and its results described. For the latter two approaches, the resulting hazard is used in the Cox model in a simulation study.^ The unconditional survival distribution formed from the first model involves dependency, but the crude hazard resulting from this unconditional distribution is identical to the marginal hazard, and inferences based on the hazard are valid. The hazard ratios formed from two distributions following the Gumbel Type A model are biased by a factor dependent on the amount of censoring in the two populations and the strength of the dependency of death and censoring in the two populations. The Cox model estimates this biased hazard ratio. In general, the hazard resulting from the compartmental model is not constant, even if the individual marginal hazards are constant, unless censoring is non-informative. The hazard ratio tends to a specific limit.^ Methods of evaluating situations in which informative censoring is present are described, and the relative utility of the three models examined is discussed. ^

Selection bias and the cross-validation of regression models for prediction

Strategies are compared for the development of a linear regression model with stochastic (multivariate normal) regressor variables and the subsequent assessment of its predictive ability. Bias and mean squared error of four estimators of predictive performance are evaluated in simulated samples of 32 population correlation matrices. Models including all of the available predictors are compared with those obtained using selected subsets. The subset selection procedures investigated include two stopping rules, C$\sb{\rm p}$ and S$\sb{\rm p}$, each combined with an 'all possible subsets' or 'forward selection' of variables. The estimators of performance utilized include parametric (MSEP$\sb{\rm m}$) and non-parametric (PRESS) assessments in the entire sample, and two data splitting estimates restricted to a random or balanced (Snee's DUPLEX) 'validation' half sample. The simulations were performed as a designed experiment, with population correlation matrices representing a broad range of data structures.^ The techniques examined for subset selection do not generally result in improved predictions relative to the full model. Approaches using 'forward selection' result in slightly smaller prediction errors and less biased estimators of predictive accuracy than 'all possible subsets' approaches but no differences are detected between the performances of C$\sb{\rm p}$ and S$\sb{\rm p}$. In every case, prediction errors of models obtained by subset selection in either of the half splits exceed those obtained using all predictors and the entire sample.^ Only the random split estimator is conditionally (on $\\beta$) unbiased, however MSEP$\sb{\rm m}$ is unbiased on average and PRESS is nearly so in unselected (fixed form) models. When subset selection techniques are used, MSEP$\sb{\rm m}$ and PRESS always underestimate prediction errors, by as much as 27 percent (on average) in small samples. Despite their bias, the mean squared errors (MSE) of these estimators are at least 30 percent less than that of the unbiased random split estimator. The DUPLEX split estimator suffers from large MSE as well as bias, and seems of little value within the context of stochastic regressor variables.^ To maximize predictive accuracy while retaining a reliable estimate of that accuracy, it is recommended that the entire sample be used for model development, and a leave-one-out statistic (e.g. PRESS) be used for assessment. ^

Penetrating injuries of the thorax and abdomen: Type of medical-care institution and case fatality

The purpose of this study was to determine, for penetrating injuries (gunshot, stab) of the chest/abdomen, the impact on fatality of treatment in trauma centers and shock trauma units compared with general hospitals. Medical records of all cases of penetrating injury limited to chest/abdomen and admitted to and discharged from 7 study facilities in Baltimore city 1979-1980 (n = 581) were studied: 4 general hospitals (n = 241), 2 area-wide trauma centers (n = 298), and a shock trauma unit (n = 42). Emergency center and transferred cases were not studied. Anatomical injury severity, measured by modified Injury Severity Score (mISS), was a significant prognostic factor for death, as were cardiovascular shock (SBP $\le$ 70), injury type (gunshot vs stab), and ambulance/helicopter (vs other) transport. All deaths occurred in cases with two or more prognostic factors. Unadjusted relative risks of death compared with general hospitals were 4.3 (95% confidence interval = 2.2, 8.4) for shock trauma and 0.8 (0.4, 1.7) for trauma centers. Controlling for prognostic factors by logistic regression resulted in these relative risks: shock trauma 4.0 (0.7, 22.2), and trauma centers 0.8 (0.2, 3.2). Factors significantly associated with increased risk had the following relative risks by multiple logistic regression: SBP $\le$ 70 (RR = 40.7 (11.0, 148.7)), highest mISS (42 (7.7, 227)), gunshot (8.4 (2.1, 32.6)), and ambulance/helicopter transport (17.2 (1.3, 228.1)). Controlling for age, race, and gender did not alter results significantly. Actual deaths compared with deaths predicted from a multivariable model of general-hospital cases showed 3.7 more than predicted deaths in shock trauma (SMR = 1.6 (0.8, 2.9)) and 0.7 more than predicted deaths in area-wide trauma centers (SMR = 1.05 (0.6, 1.7)). Selection bias due to exclusion of transfers and emergency center cases, and residual confounding due to insufficient injury information, may account for persistence of adjusted high case fatality in shock trauma. Studying all cases prospectively, including emergency center and transferred cases, is needed. ^

Assessing time-by-covariate interactions in Cox proportional hazards regression models using cubic spline functions

This dissertation develops and explores the methodology for the use of cubic spline functions in assessing time-by-covariate interactions in Cox proportional hazards regression models. These interactions indicate violations of the proportional hazards assumption of the Cox model. Use of cubic spline functions allows for the investigation of the shape of a possible covariate time-dependence without having to specify a particular functional form. Cubic spline functions yield both a graphical method and a formal test for the proportional hazards assumption as well as a test of the nonlinearity of the time-by-covariate interaction. Five existing methods for assessing violations of the proportional hazards assumption are reviewed and applied along with cubic splines to three well known two-sample datasets. An additional dataset with three covariates is used to explore the use of cubic spline functions in a more general setting. ^

A Bayesian approach to estimating the regression coefficients of a multinomial logit model

A Bayesian approach to estimation of the regression coefficients of a multinominal logit model with ordinal scale response categories is presented. A Monte Carlo method is used to construct the posterior distribution of the link function. The link function is treated as an arbitrary scalar function. Then the Gauss-Markov theorem is used to determine a function of the link which produces a random vector of coefficients. The posterior distribution of the random vector of coefficients is used to estimate the regression coefficients. The method described is referred to as a Bayesian generalized least square (BGLS) analysis. Two cases involving multinominal logit models are described. Case I involves a cumulative logit model and Case II involves a proportional-odds model. All inferences about the coefficients for both cases are described in terms of the posterior distribution of the regression coefficients. The results from the BGLS method are compared to maximum likelihood estimates of the regression coefficients. The BGLS method avoids the nonlinear problems encountered when estimating the regression coefficients of a generalized linear model. The method is not complex or computationally intensive. The BGLS method offers several advantages over Bayesian approaches. ^

Logistic regression with incompletely observed binary

Logistic regression is one of the most important tools in the analysis of epidemiological and clinical data. Such data often contain missing values for one or more variables. Common practice is to eliminate all individuals for whom any information is missing. This deletion approach does not make efficient use of available information and often introduces bias.^ Two methods were developed to estimate logistic regression coefficients for mixed dichotomous and continuous covariates including partially observed binary covariates. The data were assumed missing at random (MAR). One method (PD) used predictive distribution as weight to calculate the average of the logistic regressions performing on all possible values of missing observations, and the second method (RS) used a variant of resampling technique. Additional seven methods were compared with these two approaches in a simulation study. They are: (1) Analysis based on only the complete cases, (2) Substituting the mean of the observed values for the missing value, (3) An imputation technique based on the proportions of observed data, (4) Regressing the partially observed covariates on the remaining continuous covariates, (5) Regressing the partially observed covariates on the remaining continuous covariates conditional on response variable, (6) Regressing the partially observed covariates on the remaining continuous covariates and response variable, and (7) EM algorithm. Both proposed methods showed smaller standard errors (s.e.) for the coefficient involving the partially observed covariate and for the other coefficients as well. However, both methods, especially PD, are computationally demanding; thus for analysis of large data sets with partially observed covariates, further refinement of these approaches is needed. ^

DISTRIBUTION FUNCTIONS, MEAN SQUARED ERRORS, AND CONFIDENCE LIMITS FOR RIDGE REGRESSION ESTIMATORS

A large number of ridge regression estimators have been proposed and used with little knowledge of their true distributions. Because of this lack of knowledge, these estimators cannot be used to test hypotheses or to form confidence intervals.^ This paper presents a basic technique for deriving the exact distribution functions for a class of generalized ridge estimators. The technique is applied to five prominent generalized ridge estimators. Graphs of the resulting distribution functions are presented. The actual behavior of these estimators is found to be considerably different than the behavior which is generally assumed for ridge estimators.^ This paper also uses the derived distributions to examine the mean squared error properties of the estimators. A technique for developing confidence intervals based on the generalized ridge estimators is also presented. ^

POLYCHOTOMOUS LOGISTIC REGRESSION ANALYSIS (HYPERTENSION, CORONARY HEART DISEASE)

The history of the logistic function since its introduction in 1838 is reviewed, and the logistic model for a polychotomous response variable is presented with a discussion of the assumptions involved in its derivation and use. Following this, the maximum likelihood estimators for the model parameters are derived along with a Newton-Raphson iterative procedure for evaluation. A rigorous mathematical derivation of the limiting distribution of the maximum likelihood estimators is then presented using a characteristic function approach. An appendix with theorems on the asymptotic normality of sample sums when the observations are not identically distributed, with proofs, supports the presentation on asymptotic properties of the maximum likelihood estimators. Finally, two applications of the model are presented using data from the Hypertension Detection and Follow-up Program, a prospective, population-based, randomized trial of treatment for hypertension. The first application compares the risk of five-year mortality from cardiovascular causes with that from noncardiovascular causes; the second application compares risk factors for fatal or nonfatal coronary heart disease with those for fatal or nonfatal stroke. ^

REGRESSION ANALYSIS FOR STANDARDIZED MORTALITY RATIOS WITH APPLICATION TO OCCUPATIONAL FOLLOW-UP STUDIES

Traditional comparison of standardized mortality ratios (SMRs) can be misleading if the age-specific mortality ratios are not homogeneous. For this reason, a regression model has been developed which incorporates the mortality ratio as a function of age. This model is then applied to mortality data from an occupational cohort study. The nature of the occupational data necessitates the investigation of mortality ratios which increase with age. These occupational data are used primarily to illustrate and develop the statistical methodology.^ The age-specific mortality ratio (MR) for the covariates of interest can be written as MR(,ij...m) = ((mu)(,ij...m)/(theta)(,ij...m)) = r(.)exp (Z('')(,ij...m)(beta)) where (mu)(,ij...m) and (theta)(,ij...m) denote the force of mortality in the study and chosen standard populations in the ij...m('th) stratum, respectively, r is the intercept, Z(,ij...m) is the vector of covariables associated with the i('th) age interval, and (beta) is a vector of regression coefficients associated with these covariables. A Newton-Raphson iterative procedure has been used for determining the maximum likelihood estimates of the regression coefficients.^ This model provides a statistical method for a logical and easily interpretable explanation of an occupational cohort mortality experience. Since it gives a reasonable fit to the mortality data, it can also be concluded that the model is fairly realistic. The traditional statistical method for the analysis of occupational cohort mortality data is to present a summary index such as the SMR under the assumption of constant (homogeneous) age-specific mortality ratios. Since the mortality ratios for occupational groups usually increase with age, the homogeneity assumption of the age-specific mortality ratios is often untenable. The traditional method of comparing SMRs under the homogeneity assumption is a special case of this model, without age as a covariate.^ This model also provides a statistical technique to evaluate the relative risk between two SMRs or a dose-response relationship among several SMRs. The model presented has application in the medical, demographic and epidemiologic areas. The methods developed in this thesis are suitable for future analyses of mortality or morbidity data when the age-specific mortality/morbidity experience is a function of age or when there is an interaction effect between confounding variables needs to be evaluated. ^