A flexible approach for causal inference with multiple treatments and clustered survival outcomes.

When drawing causal inferences about the effects of multiple treatments on clustered survival outcomes using observational data, we need to address implications of the multilevel data structure, multiple treatments, censoring, and unmeasured confounding for causal analyses. Few off-the-shelf causal inference tools are available to simultaneously tackle these issues. We develop a flexible random-intercept accelerated failure time model, in which we use Bayesian additive regression trees to capture arbitrarily complex relationships between censored survival times and pre-treatment covariates and use the random intercepts to capture cluster-specific main effects. We develop an efficient Markov chain Monte Carlo algorithm to draw posterior inferences about the population survival effects of multiple treatments and examine the variability in cluster-level effects. We further propose an interpretable sensitivity analysis approach to evaluate the sensitivity of drawn causal inferences about treatment effect to the potential magnitude of departure from the causal assumption of no unmeasured confounding. Expansive simulations empirically validate and demonstrate good practical operating characteristics of our proposed methods. Applying the proposed methods to a dataset on older high-risk localized prostate cancer patients drawn from the National Cancer Database, we evaluate the comparative effects of three treatment approaches on patient survival, and assess the ramifications of potential unmeasured confounding. The methods developed in this work are readily available in the R $$ \mathsf{R}\kern.15em $$ package riAFTBART $$ \mathsf{riAFTBART} $$ .

Integrated community-based HIV and non-communicable disease care within microfinance groups in Kenya: study protocol for the Harambee cluster randomised trial.

INTRODUCTION: In Kenya, distance to health facilities, inefficient vertical care delivery and limited financial means are barriers to retention in HIV care. Furthermore, the increasing burden of non-communicable diseases (NCDs) among people living with HIV complicates chronic disease treatment and strains traditional care delivery models. Potential strategies for improving HIV/NCD treatment outcomes are differentiated care, community-based care and microfinance (MF). METHODS AND ANALYSIS: We will use a cluster randomised trial to evaluate integrated community-based (ICB) care incorporated into MF groups in medium and high HIV prevalence areas in western Kenya. We will conduct baseline assessments with n=900 HIV positive members of 40 existing MF groups. Group clusters will be randomised to receive either (1) ICB or (2) standard of care (SOC). The ICB intervention will include: (1) clinical care visits during MF group meetings inclusive of medical consultations, NCD management, distribution of antiretroviral therapy (ART) and NCD medications, and point-of-care laboratory testing; (2) peer support for ART adherence and (3) facility referrals as needed. MF groups randomised to SOC will receive regularly scheduled care at a health facility. Findings from the two trial arms will be compared with follow-up data from n=300 matched controls. The primary outcome will be VS at 18 months. Secondary outcomes will be retention in care, absolute mean change in systolic blood pressure and absolute mean change in HbA1c level at 18 months. We will use mediation analysis to evaluate mechanisms through which MF and ICB care impact outcomes and analyse incremental cost-effectiveness of the intervention in terms of cost per HIV suppressed person-time, cost per patient retained in care and cost per disability-adjusted life-year saved. ETHICS AND DISSEMINATION: The Moi University Institutional Research and Ethics Committee approved this study (IREC#0003054). We will share data via the Brown University Digital Repository and disseminate findings via publication. TRIAL REGISTRATION NUMBER: NCT04417127.

Predictors of Health Insurance Enrollment among HIV Positive Pregnant Women in Kenya: Potential for Adverse Selection and Implications for HIV Treatment and Prevention.

Background: The global push to achieve the 90-90-90 targets designed to end the HIV epidemic has called for the removing of policy barriers to prevention and treatment, and ensuring financial sustainability of HIV programs. Universal health insurance is one tool that can be used to this end. In sub-Saharan Africa, where HIV prevalence and incidence remain high, the use of health insurance to provide comprehensive HIV care is limited. This study looked at the factors that best predict social health insurance enrollment among HIV positive pregnant women using data from the Academic Model Providing Access to Healthcare (AMPATH) in western Kenya. Methods: Cross-sectional clinical encounter data were extracted from the electronic medical records (EMR) at AMPATH. We used univariate and multivariate logistic regressions to estimate the predictors of health insurance enrollment among HIV positive pregnant women. The analysis was further stratified by HIV disease severity (based on CD4 cell count <350 and 350>) to test the possibility of differential enrollment given HIV disease state. Results: Approximately 7% of HIV infected women delivering at a healthcare facility had health insurance. HIV positive pregnant women who deliver at a health facility had twice the odds of enrolling in insurance [2.46 Adjusted Odds Ratio (AOR), Confidence Interval (CI) 1.24-4.87]. They were 10 times more likely to have insurance if they were lost to follow-up to HIV care during pregnancy [9.90 AOR; CI 3.42-28.67], and three times more likely to enroll if they sought care at an urban clinic [2.50 AOR; 95% CI 1.53-4.12]. Being on HIV treatment was negatively associated with health insurance enrollment [0.22 AOR; CI 0.10-0.49]. Stratifying the analysis by HIV disease severity while statistically significant did not change these results. Conclusions: The findings indicated that health insurance enrollment among HIV positive pregnant women was low mirroring national levels. Additionally, structural factors, such as access to institutional delivery and location of healthcare facilities, increased the likelihood of health insurance enrollment within this population. However, behavioral aspects, such as being lost to follow-up to HIV care during pregnancy and being on HIV treatment, had an ambiguous effect on insurance enrollment. This may potentially be because of adverse selection and information asymmetries. Further understanding of the relationship between insurance and HIV is needed if health insurance is to be utilized for HIV treatment and prevention in limited resource settings.

Evaluating the Population Impact on Racial/Ethnic Disparities in HIV in Adulthood of Intervening on Specific Targets: A Conceptual and Methodological Framework.

Reducing racial/ethnic disparities in human immunodeficiency virus (HIV) disease is a high priority. Reductions in HIV racial/ethnic disparities can potentially be achieved by intervening on important intermediate factors. The potential population impact of intervening on intermediates can be evaluated using observational data when certain conditions are met. However, using standard stratification-based approaches commonly employed in the observational HIV literature to estimate the potential population impact in this setting may yield results that do not accurately estimate quantities of interest. Here we describe a useful conceptual and methodological framework for using observational data to appropriately evaluate the impact on HIV racial/ethnic disparities of interventions. This framework reframes relevant scientific questions in terms of a controlled direct effect and estimates a corresponding proportion eliminated. We review methods and conditions sufficient for accurate estimation within the proposed framework. We use the framework to analyze data on 2,329 participants in the CFAR [Centers for AIDS Research] Network of Integrated Clinical Systems (2008-2014) to evaluate the potential impact of universal prescription of and ≥95% adherence to antiretroviral therapy on racial disparities in HIV virological suppression. We encourage the use of the described framework to appropriately evaluate the potential impact of targeted interventions in addressing HIV racial/ethnic disparities using observational data.

A state transition framework for patient-level modeling of engagement and retention in HIV care using longitudinal cohort data.

The human immunodeficiency virus (HIV) care cascade is a conceptual model used to outline the benchmarks that reflects effectiveness of HIV care in the whole HIV care continuum. The models can be used to identify barriers contributing to poor outcomes along each benchmark in the cascade such as disengagement from care or death. Recently, the HIV care cascade has been widely applied to monitor progress towards HIV prevention and care goals in an attempt to develop strategies to improve health outcomes along the care continuum. Yet, there are challenges in quantifying successes and gaps in HIV care using the cascade models that are partly due to the lack of analytic approaches. The availability of large cohort data presents an opportunity to develop a coherent statistical framework for analysis of the HIV care cascade. Motivated by data from the Academic Model Providing Access to Healthcare, which has provided HIV care to nearly 200,000 individuals in Western Kenya since 2001, we developed a state transition framework that can characterize patient-level movements through the multiple stages of the HIV care cascade. We describe how to transform large observational data into an analyzable format. We then illustrate the state transition framework via multistate modeling to quantify dynamics in retention aspects of care. The proposed modeling approach identifies the transition probabilities of moving through each stage in the care cascade. In addition, this approach allows regression-based estimation to characterize effects of (time-varying) predictors of within and between state transitions such as retention, disengagement, re-entry into care, transfer-out, and mortality. Copyright © 2017 John Wiley & Sons, Ltd.

Bridging Income Generation with Group Integrated Care for cardiovascular risk reduction: Rationale and design of the BIGPIC study.

BACKGROUND: Cardiovascular disease (CVD) is the leading cause of mortality worldwide, with >80% of CVD deaths occurring in low and middle income countries (LMICs). Diabetes mellitus and pre-diabetes are risk factors for CVD, and CVD is the major cause of morbidity and mortality among individuals with DM. There is a critical period now during which reducing CVD risk among individuals with diabetes and pre-diabetes may have a major impact. Cost-effective, culturally appropriate, and context-specific approaches are required. Two promising strategies to improve health outcomes are group medical visits and microfinance. METHODS/DESIGN: This study tests whether group medical visits integrated into microfinance groups are effective and cost-effective in reducing CVD risk among individuals with diabetes or at increased risk for diabetes in western Kenya. An initial phase of qualitative inquiry will assess contextual factors, facilitators, and barriers that may impact integration of group medical visits and microfinance for CVD risk reduction. Subsequently, we will conduct a four-arm cluster randomized trial comparing: (1) usual clinical care, (2) usual clinical care plus microfinance groups only, (3) group medical visits only, and (4) group medical visits integrated into microfinance groups. The primary outcome measure will be 1-year change in systolic blood pressure, and a key secondary outcome measure is 1-year change in overall CVD risk as measured by the QRISK2 score. We will conduct mediation analysis to evaluate the influence of changes in social network characteristics on intervention outcomes, as well as moderation analysis to evaluate the influence of baseline social network characteristics on effectiveness of the interventions. Cost-effectiveness analysis will be conducted in terms of cost per unit change in systolic blood pressure, percent change in CVD risk score, and per disability-adjusted life year saved. DISCUSSION: This study will provide evidence regarding effectiveness and cost-effectiveness of interventions to reduce CVD risk. We aim to produce generalizable methods and results that can provide a model for adoption in low-resource settings worldwide.

Evaluating strategies to improve HIV care outcomes in Kenya: a modelling study.

BACKGROUND: With expanded access to antiretroviral therapy (ART) in sub-Saharan Africa, HIV mortality has decreased, yet life-years are still lost to AIDS. Strengthening of treatment programmes is a priority. We examined the state of an HIV care programme in Kenya and assessed interventions to improve the impact of ART programmes on population health. METHODS: We created an individual-based mathematical model to describe the HIV epidemic and the experiences of care among adults infected with HIV in Kenya. We calibrated the model to a longitudinal dataset from the Academic Model Providing Access To Healthcare (known as AMPATH) programme describing the routes into care, losses from care, and clinical outcomes. We simulated the cost and effect of interventions at different stages of HIV care, including improvements to diagnosis, linkage to care, retention and adherence of ART, immediate ART eligibility, and a universal test-and-treat strategy. FINDINGS: We estimate that, of people dying from AIDS between 2010 and 2030, most will have initiated treatment (61%), but many will never have been diagnosed (25%) or will have been diagnosed but never started ART (14%). Many interventions targeting a single stage of the health-care cascade were likely to be cost-effective, but any individual intervention averted only a small percentage of deaths because the effect is attenuated by other weaknesses in care. However, a combination of five interventions (including improved linkage, point-of-care CD4 testing, voluntary counselling and testing with point-of-care CD4, and outreach to improve retention in pre-ART care and on-ART) would have a much larger impact, averting 1·10 million disability-adjusted life-years (DALYs) and 25% of expected new infections and would probably be cost-effective (US$571 per DALY averted). This strategy would improve health more efficiently than a universal test-and-treat intervention if there were no accompanying improvements to care ($1760 per DALY averted). INTERPRETATION: When resources are limited, combinations of interventions to improve care should be prioritised over high-cost strategies such as universal test-and-treat strategy, especially if this is not accompanied by improvements to the care cascade. International guidance on ART should reflect alternative routes to programme strengthening and encourage country programmes to evaluate the costs and population-health impact in addition to the clinical benefits of immediate initiation. FUNDING: Bill & Melinda Gates Foundation, United States Agency for International Development, National Institutes of Health.

Optimizing linkage and retention to hypertension care in rural Kenya (LARK hypertension study): study protocol for a randomized controlled trial.

BACKGROUND: Hypertension is the leading global risk factor for mortality. Hypertension treatment and control rates are low worldwide, and delays in seeking care are associated with increased mortality. Thus, a critical component of hypertension management is to optimize linkage and retention to care. METHODS/DESIGN: This study investigates whether community health workers, equipped with a tailored behavioral communication strategy and smartphone technology, can increase linkage and retention of hypertensive individuals to a hypertension care program and significantly reduce blood pressure among them. The study will be conducted in the Kosirai and Turbo Divisions of western Kenya. An initial phase of qualitative inquiry will assess facilitators and barriers of linkage and retention to care using a modified Health Belief Model as a conceptual framework. Subsequently, we will conduct a cluster randomized controlled trial with three arms: 1) usual care (community health workers with the standard level of hypertension care training); 2) community health workers with an additional tailored behavioral communication strategy; and 3) community health workers with a tailored behavioral communication strategy who are also equipped with smartphone technology. The co-primary outcome measures are: 1) linkage to hypertension care, and 2) one-year change in systolic blood pressure among hypertensive individuals. Cost-effectiveness analysis will be conducted in terms of costs per unit decrease in blood pressure and costs per disability-adjusted life year gained. DISCUSSION: This study will provide evidence regarding the effectiveness and cost-effectiveness of strategies to optimize linkage and retention to hypertension care that can be applicable to non-communicable disease management in low- and middle-income countries. TRIAL REGISTRATION: This trial is registered with (NCT01844596) on 30 April 2013.

Optimal Allocation of Gold Standard Testing under Constrained Availability: Application to Assessment of HIV Treatment Failure.

The World Health Organization (WHO) guidelines for monitoring the effectiveness of HIV treatment in resource-limited settings (RLS) are mostly based on clinical and immunological markers (e.g., CD4 cell counts). Recent research indicates that the guidelines are inadequate and can result in high error rates. Viral load (VL) is considered the "gold standard", yet its widespread use is limited by cost and infrastructure. In this paper, we propose a diagnostic algorithm that uses information from routinely-collected clinical and immunological markers to guide a selective use of VL testing for diagnosing HIV treatment failure, under the assumption that VL testing is available only at a certain portion of patient visits. Our algorithm identifies the patient sub-population, such that the use of limited VL testing on them minimizes a pre-defined risk (e.g., misdiagnosis error rate). Diagnostic properties of our proposal algorithm are assessed by simulations. For illustration, data from the Miriam Hospital Immunology Clinic (RI, USA) are analyzed.

Varying-coefficient models for longitudinal processes with continuous-time informative dropout.

Dropout is a common occurrence in longitudinal studies. Building upon the pattern-mixture modeling approach within the Bayesian paradigm, we propose a general framework of varying-coefficient models for longitudinal data with informative dropout, where measurement times can be irregular and dropout can occur at any point in continuous time (not just at observation times) together with administrative censoring. Specifically, we assume that the longitudinal outcome process depends on the dropout process through its model parameters. The unconditional distribution of the repeated measures is a mixture over the dropout (administrative censoring) time distribution, and the continuous dropout time distribution with administrative censoring is left completely unspecified. We use Markov chain Monte Carlo to sample from the posterior distribution of the repeated measures given the dropout (administrative censoring) times; Bayesian bootstrapping on the observed dropout (administrative censoring) times is carried out to obtain marginal covariate effects. We illustrate the proposed framework using data from a longitudinal study of depression in HIV-infected women; the strategy for sensitivity analysis on unverifiable assumption is also demonstrated.

Step into Motion: a randomized trial examining the relative efficacy of Internet vs. print-based physical activity interventions.

Over two-thirds of Americans access the Internet and therefore, the Internet may be an important channel for reaching the large population of sedentary individuals. The purpose of this paper is to describe the methods for a randomized controlled trial examining the efficacy of an Internet-based physical activity intervention relative to a print intervention that has been shown to be effective in previous trials. Specifically, 249 sedentary participants were randomized to receive one of three interventions: 1) Internet-based motivationally-tailored individualized feedback (Tailored Internet); 2) print-based motivationally-tailored individualized feedback (Tailored Print); or 3) physical activity websites currently available to the public (Standard Internet). Participants completed the 7-Day Physical Activity Recall interview, wore an objective physical activity monitor (i.e., ActiGraph), and participated in a treadmill fitness test at baseline, 6, and 12 months. The sample consisted of mostly women (84.2%) and Caucasian individuals (76.4%) who reported exercising an average of 21 min per week at baseline. This is the first study that we are aware of, that has examined the efficacy of a tailored Internet-based physical activity intervention. This study will have implications for the dissemination of Internet-based physical activity interventions.