Michael Schomaker

Life Expectancies of South African Adults Starting Antiretroviral Treatment: Collaborative Analysis of Cohort Studies

Leigh Johnson and colleagues estimate the life expectancies of HIV positive South African adults who are taking antiretroviral therapy by using information from 6 programmes between 2001 and 2010.

Gender differences in survival among adult patients starting antiretroviral therapy in South Africa: a multicentre cohort study.

Morna Cornell and colleagues investigate differences in mortality for HIV-positive men and women on antiretroviral therapy in South Africa.

Effectiveness of Patient Adherence Groups as a Model of Care for Stable Patients on Antiretroviral Therapy in Khayelitsha, Cape Town, South Africa

Background Innovative models of care are required to cope with the ever-increasing number of patients on antiretroviral therapy in the most affected countries. This study, in Khayelitsha, South Africa, evaluates the effectiveness of a group-based model of care run predominantly by non-clinical staff in retaining patients in care and maintaining adherence. Methods and Findings Participation in “adherence clubs” was offered to adults who had been on ART for at least 18 months, had a current CD4 count >200 cells/ml and were virologically suppressed. Embedded in an ongoing cohort study, we compared loss to care and virologic rebound in patients receiving the intervention with patients attending routine nurse-led care from November 2007 to February 2011. We used inverse probability weighting to estimate the intention-to-treat effect of adherence club participation, adjusted for measured baseline and time-varying confounders. The principal outcome was the combination of death or loss to follow-up. The secondary outcome was virologic rebound in patients who were virologically suppressed at study entry. Of 2829 patients on ART for >18 months with a CD4 count above 200 cells/µl, 502 accepted club participation. At the end of the study, 97% of club patients remained in care compared with 85% of other patients. In adjusted analyses club participation reduced loss-to-care by 57% (hazard ratio [HR] 0.43, 95% CI = 0.21–0.91) and virologic rebound in patients who were initially suppressed by 67% (HR 0.33, 95% CI = 0.16–0.67). Discussion Patient adherence groups were found to be an effective model for improving retention and documented virologic suppression for stable patients in long term ART care. Out-of-clinic group-based models facilitated by non-clinical staff are a promising approach to assist in the long-term management of people on ART in high burden low or middle-income settings.

Frequentist Model Averaging with missing observations

Model averaging or combining is often considered as an alternative to model selection. Frequentist Model Averaging (FMA) is considered extensively and strategies for the application of FMA methods in the presence of missing data based on two distinct approaches are presented. The first approach combines estimates from a set of appropriate models which are weighted by scores of a missing data adjusted criterion developed in the recent literature of model selection. The second approach averages over the estimates of a set of models with weights based on conventional model selection criteria but with the missing data replaced by imputed values prior to estimating the models. For this purpose three easy-to-use imputation methods that have been programmed in currently available statistical software are considered, and a simple recursive algorithm is further adapted to implement a generalized regression imputation in a way such that the missing values are predicted successively. The latter algorithm is found to be quite useful when one is confronted with two or more missing values simultaneously in a given row of observations. Focusing on a binary logistic regression model, the properties of the FMA estimators resulting from these strategies are explored by means of a Monte Carlo study. The results show that in many situations, averaging after imputation is preferred to averaging using weights that adjust for the missing data, and model average estimators often provide better estimates than those resulting from any single model. As an illustration, the proposed methods are applied to a dataset from a study of Duchenne muscular dystrophy detection.

Model selection and model averaging after multiple imputation

Model selection and model averaging are two important techniques to obtain practical and useful models in applied research. However, it is now well-known that many complex issues arise, especially in the context of model selection, when the stochastic nature of the selection process is ignored and estimates, standard errors, and confidence intervals are calculated as if the selected model was known a priori. While model averaging aims to incorporate the uncertainty associated with the model selection process by combining estimates over a set of models, there is still some debate over appropriate interpretation and confidence interval construction. These problems become even more complex in the presence of missing data and it is currently not entirely clear how to proceed. To deal with such situations, a framework for model selection and model averaging in the context of missing data is proposed. The focus lies on multiple imputation as a strategy to deal with the missingness: a consequent combination with model averaging aims to incorporate both the uncertainty associated with the model selection and with the imputation process. Furthermore, the performance of bootstrapping as a flexible extension to our framework is evaluated. Monte Carlo simulations are used to reveal the nature of the proposed estimators in the context of the linear regression model. The practical implications of our approach are illustrated by means of a recent survival study on sputum culture conversion in pulmonary tuberculosis.

Baseline predictors of sputum culture conversion in pulmonary tuberculosis: Importance of cavities, smoking, time to detection and w-beijing genotype

Background Time to detection (TTD) on automated liquid mycobacterial cultures is an emerging biomarker of tuberculosis outcomes. The M. tuberculosis W-Beijing genotype is spreading globally, indicating a selective advantage. There is a paucity of data on the association between baseline TTD and W-Beijing genotype and tuberculosis outcomes. Aim To assess baseline predictors of failure of sputum culture conversion, within the first 2 months of antitubercular therapy, in participants with pulmonary tuberculosis. Design Between May 2005 and August 2008 we conducted a prospective cohort study of time to sputum culture conversion in ambulatory participants with first episodes of smear and culture positive pulmonary tuberculosis attending two primary care clinics in Cape Town, South Africa. Rifampicin resistance (diagnosed on phenotypic susceptibility testing) was an exclusion criterion. Sputum was collected weekly for 8 weeks for mycobacterial culture on liquid media (BACTEC MGIT 960). Due to missing data, multiple imputation was performed. Time to sputum culture conversion was analysed using a Cox-proportional hazards model. Bayesian model averaging determined the posterior effect probability for each variable. Results 113 participants were enrolled (30.1% female, 10.5% HIV-infected, 44.2% W-Beijing genotype, and 89% cavities). On Kaplan Meier analysis 50.4% of participants underwent sputum culture conversion by 8 weeks. The following baseline factors were associated with slower sputum culture conversion: TTD (adjusted hazard ratio (aHR) = 1.11, 95% CI 1.02; 1.2), lung cavities (aHR = 0.13, 95% CI 0.02; 0.95), ever smoking (aHR = 0.32, 95% CI 0.1; 1.02) and the W-Beijing genotype (aHR = 0.51, 95% CI 0.25; 1.07). On Bayesian model averaging, posterior probability effects were strong for TTD, lung cavitation and smoking and moderate for W-Beijing genotype. Conclusion We found that baseline TTD, smoking, cavities and W-Beijing genotype were associated with delayed 2 month sputum culture. Larger studies are needed to confirm the relationship between the W-Beijing genotype and sputum culture conversion.

Targeted Maximum Likelihood Estimation for Dynamic and Static Longitudinal Marginal Structural Working Models.

Abstract This paper describes a targeted maximum likelihood estimator (TMLE) for the parameters of longitudinal static and dynamic marginal structural models. We consider a longitudinal data structure consisting of baseline covariates, time-dependent intervention nodes, intermediate time-dependent covariates, and a possibly time-dependent outcome. The intervention nodes at each time point can include a binary treatment as well as a right-censoring indicator. Given a class of dynamic or static interventions, a marginal structural model is used to model the mean of the intervention-specific counterfactual outcome as a function of the intervention, time point, and possibly a subset of baseline covariates. Because the true shape of this function is rarely known, the marginal structural model is used as a working model. The causal quantity of interest is defined as the projection of the true function onto this working model. Iterated conditional expectation double robust estimators for marginal structural model parameters were previously proposed by Robins (2000, 2002) and Bang and Robins (2005). Here we build on this work and present a pooled TMLE for the parameters of marginal structural working models. We compare this pooled estimator to a stratified TMLE (Schnitzer et al. 2014) that is based on estimating the intervention-specific mean separately for each intervention of interest. The performance of the pooled TMLE is compared to the performance of the stratified TMLE and the performance of inverse probability weighted (IPW) estimators using simulations. Concepts are illustrated using an example in which the aim is to estimate the causal effect of delayed switch following immunological failure of first line antiretroviral therapy among HIV-infected patients. Data from the International Epidemiological Databases to Evaluate AIDS, Southern Africa are analyzed to investigate this question using both TML and IPW estimators. Our results demonstrate practical advantages of the pooled TMLE over an IPW estimator for working marginal structural models for survival, as well as cases in which the pooled TMLE is superior to its stratified counterpart.

Effects of rifampin-based antituberculosis therapy on plasma efavirenz concentrations in children vary by CYP2B6 genotype

Objectives:An efavirenz-based antiretroviral therapy (ART) regimen is preferred for children more than 3 years of age with tuberculosis. However, rifampin, a key component of antituberculosis therapy, induces CYP2B6. An increased dose of efavirenz is recommended in adults weighing more than 50 kg who require rifampin, but there is scant information in children being treated for tuberculosis. Design:Plasma efavirenz concentrations were compared in 40 children during concomitant treatment for tuberculosis and HIV-1, after stopping rifampicin, and in a control group of children without tuberculosis. Associations with antituberculosis treatment, metabolizer genotype (based on CYP2B6 516G→T, 983T→C, and 15582C→T), weight, and time after dose were evaluated. Results:Compared to children with extensive metabolizer genotypes, efavirenz concentrations were increased 1.42-fold (95% confidence interval, CI 0.94–2.15) and 2.85-fold (95% CI 1.80–4.52) in children with intermediate and slow metabolizer genotypes, respectively. Concomitant antituberculosis treatment increased efavirenz concentrations 1.49-fold (95% CI 1.10–2.01) in children with slow metabolizer genotypes, but did not affect efavirenz concentrations in extensive or intermediate metabolizer genotypes. After adjustment for dose/kg, each kilogram of weight was associated with a 2.8% (95% CI 0.9–4.7) decrease in efavirenz concentrations. Despite higher milligram per kilogram doses, a higher proportion of children in the lowest weight band (10–13.9 kg) had efavirenz concentrations less than 1.0 mg/l than larger children. Conclusion:Antituberculosis treatment was not associated with reduced efavirenz concentrations in children, which does not support increased efavirenz doses. Children with slow metabolizer genotype have increased efavirenz concentrations during antituberculosis treatment, likely due to isoniazid inhibiting enzymes involved in accessory metabolic pathways for efavirenz.

Non-ignorable loss to follow-up: correcting mortality estimates based on additional outcome ascertainment

Loss to follow-up (LTFU) is a common problem in many epidemiological studies. In antiretroviral treatment (ART) programs for patients with human immunodeficiency virus (HIV), mortality estimates can be biased if the LTFU mechanism is non-ignorable, that is, mortality differs between lost and retained patients. In this setting, routine procedures for handling missing data may lead to biased estimates. To appropriately deal with non-ignorable LTFU, explicit modeling of the missing data mechanism is needed. This can be based on additional outcome ascertainment for a sample of patients LTFU, for example, through linkage to national registries or through survey-based methods. In this paper, we demonstrate how this additional information can be used to construct estimators based on inverse probability weights (IPW) or multiple imputation. We use simulations to contrast the performance of the proposed estimators with methods widely used in HIV cohort research for dealing with missing data. The practical implications of our approach are illustrated using South African ART data, which are partially linkable to South African national vital registration data. Our results demonstrate that while IPWs and proper imputation procedures can be easily constructed from additional outcome ascertainment to obtain valid overall estimates, neglecting non-ignorable LTFU can result in substantial bias. We believe the proposed estimators are readily applicable to a growing number of studies where LTFU is appreciable, but additional outcome data are available through linkage or surveys of patients LTFU.

Virologic Response in Children Treated with Abacavir Compared with Stavudine-Based Antiretroviral Treatment – A South African Multi-Cohort Analysis

Background: Initiation criteria and pediatric antiretroviral treatment regimens have changed over the past few years in South Africa. We reported worse early virological outcomes associated with the use of abacavir (ABC)-based regimens at 1 large site: here, we expand this analysis to multiple sites in the IeDEA-Southern Africa collaboration. Methods: Data for 9543 antiretroviral treatment-naïve children <16 years at treatment initiation started on either stavudine/lamivudine (d4T/3TC) or ABC/3TC with efavirenz (EFV) or ritonavir-boosted lopinavir (LPV/r) treated at 6 clinics in Johannesburg and Cape Town, South Africa, were analyzed with &khgr;2 tests and logistic regression to evaluate viral suppression at 6 and 12 months. Results: Prevalence of viral suppression at 6 months in 2174 children started on a d4T-based LPV/r regimen was greater (70%) than among 438 children started on an ABC-based LPV/r regimen (54%, P < 0.0001). Among 3189 children started on a d4T-based EFV regimen, a higher proportion (86%) achieved suppression at 6 months compared with 391 children started on ABC-containing EFV regimens (78%, P < 0.0001). Relative benefit of d4T versus ABC on 6-month suppression remained in multivariate analysis after adjustment for pretreatment characteristics, cohort and year of program [LPV/r: odds ratio = 0.57 (confidence interval: 0.46–0.72); EFV: odds ratio = 0.46 (confidence interval: 0.32–0.65)]. Conclusions: This expanded analysis is consistent with our previous report of worse virological outcomes after ABC was introduced as part of first-line antiretroviral treatment in South Africa. Whether due to the drug itself or coincident with other changes over time, continued monitoring and analyses must clarify causes and prevent suboptimal long-term outcomes.