Investigation of the structure and magnitude of time-varying uncontrolled confounding in simulated cohort data analyzed using g-computation

BACKGROUND

When estimating the effect of time-varying exposures on longer-term outcomes, the assumption of conditional exchangeability or no uncontrolled confounding extends beyond baseline confounding to include time-varying confounding. We illustrate the structures and magnitude of uncontrolled time-varying confounding in exposure effect estimates obtained from g-computation when sequential conditional exchangeability is violated.

METHODS

We used directed acyclic graphs (DAGs) to depict time-varying uncontrolled confounding. We performed simulations and used g-computation to quantify the effects of each time-varying exposure for each DAG type. Models adjusting all time-varying confounders were considered the true (bias-adjusted) estimate. The exclusion of time-varying uncontrolled confounders represented the biased effect estimate and an unmet 'no uncontrolled confounding' assumption. True and biased estimates were compared across DAGs, with different magnitudes of uncontrolled confounding.

RESULTS

Time-varying uncontrolled confounding can present in several scenarios, including relationships into subsequently measured exposure(s), outcome, unmeasured confounder(s) and other measured confounder(s). In simulations, effect estimates obtained from g-computation were more biased in DAGs when the uncontrolled confounders were directly related to the outcome. Complex DAGs that included relationships between uncontrolled confounders and other variables and relationships where exposures caused uncontrolled confounders at the next time point resulted in the most biased effect estimates. In these complex DAGs, excluding uncontrolled confounders affected the multiple effect estimates.

CONCLUSIONS

Time-varying uncontrolled confounding has the potential to substantially impact observed effect estimates. Given the importance of longitudinal studies in advising public health, the impact of time-varying uncontrolled confounding warrants more recognition and evaluation using quantitative bias analysis.

Observational studies of treatment effectiveness in neurology

The capacity and power of data from cohorts, registries and randomized trials to provide answers to contemporary clinical questions in neurology has increased considerably over the past two decades. Novel sophisticated statistical methods are enabling us to harness these data to guide treatment decisions, but their complexity is making appraisal of clinical evidence increasingly demanding. In this review, we discuss several methodological aspects of contemporary research of treatment effectiveness in observational data in neurology, aimed at academic neurologists and analysts specializing in outcomes research. The review discusses specifics of the sources of observational data and their key features. It focuses on the limitations of observational data and study design, as well as statistical approaches aimed to overcome these limitations. Among the examples of leading clinical themes typically studied with analyses of observational data, the review discusses methodological approaches to comparative treatment effectiveness, development of diagnostic criteria and definitions of clinical outcomes. Finally, this review provides a brief summary of key points that will help clinical audience critically evaluate design and analytical aspects of studies of disease outcomes using observational data.

A marginal structural model to estimate the effect of antidepressant medication treatment on major cardiovascular events among people with post-traumatic stress disorder

BACKGROUND

Previous evidence on antidepressant medication and cardiovascular disease (CVD) among patients with posttraumatic stress disorder (PTSD) has been inconclusive. We estimated the association between antidepressant medication and CVD by applying a marginal structural model.

METHODS

We analyzed medical utilization records of 27 170 people with PTSD without prior major cardiovascular events in the Korean National Health Insurance Database (NHID). PTSD and CVD were defined in accordance with the recorded ICD-10 diagnostic codes. We acquired information on antidepressant use from the NHID and categorized them by medication type. A composite major adverse cardiovascular events (MACE) outcome was defined as coronary artery disease with revascularization, ischaemic stroke, and/or haemorrhagic stroke. We used inverse probability of treatment weighting to estimate the parameters of a marginal structural discrete-time survival analysis regression model, comparing the resulting estimates to those derived from traditional time-fixed and time-varying Cox proportional hazards regression. We calculated cumulative daily defined doses to test for a dose-response relationship.

RESULTS

People exposed to antidepressants showed a higher hazard of MACE [hazard ratio (HR) 1.34, 95% confidence interval (CI) 1.18-1.53]. The estimated effects were strongest for selective serotonin reuptake inhibitors (HR 1.24, 95% CI 1.08-1.44) and TCAs (HR 1.33, 95% CI 1.13-1.56). Exposure to serotonin-norepinephrine reuptake inhibitors did not appear to increase the risk of MACE. People exposed to higher doses of antidepressants showed higher risk of MACE.

CONCLUSIONS

In a national cohort of people with PTSD, exposure to antidepressant medications increased the risk of MACE in a dose-response fashion.

Within-Person Variability Score-Based Causal Inference: A Two-Step Estimation for Joint Effects of Time-Varying Treatments

Behavioral science researchers have shown strong interest in disaggregating within-person relations from between-person differences (stable traits) using longitudinal data. In this paper, we propose a method of within-person variability score-based causal inference for estimating joint effects of time-varying continuous treatments by controlling for stable traits of persons. After explaining the assumed data-generating process and providing formal definitions of stable trait factors, within-person variability scores, and joint effects of time-varying treatments at the within-person level, we introduce the proposed method, which consists of a two-step analysis. Within-person variability scores for each person, which are disaggregated from stable traits of that person, are first calculated using weights based on a best linear correlation preserving predictor through structural equation modeling (SEM). Causal parameters are then estimated via a potential outcome approach, either marginal structural models (MSMs) or structural nested mean models (SNMMs), using calculated within-person variability scores. Unlike the approach that relies entirely on SEM, the present method does not assume linearity for observed time-varying confounders at the within-person level. We emphasize the use of SNMMs with G-estimation because of its property of being doubly robust to model misspecifications in how observed time-varying confounders are functionally related to treatments/predictors and outcomes at the within-person level. Through simulation, we show that the proposed method can recover causal parameters well and that causal estimates might be severely biased if one does not properly account for stable traits. An empirical application using data regarding sleep habits and mental health status from the Tokyo Teen Cohort study is also provided.

Comparative Effectiveness of Dynamic Treatment Strategies for Medication Use and Dosage: Emulating a Target Trial Using Observational Data

BACKGROUND

Availability of detailed data from electronic health records (EHRs) has increased the potential to examine the comparative effectiveness of dynamic treatment strategies using observational data. Inverse probability (IP) weighting of dynamic marginal structural models can control for time-varying confounders. However, IP weights for continuous treatments may be sensitive to model choice.

METHODS

We describe a target trial comparing strategies for treating anemia with darbepoetin in hemodialysis patients using EHR data from the UK Renal Registry 2004 to 2016. Patients received a specified dose (microgram/week) or did not receive darbepoetin. We compared 4 methods for modeling time-varying treatment: (A) logistic regression for zero dose, standard linear regression for log dose; (B) logistic regression for zero dose, heteroscedastic linear regression for log dose; (C) logistic regression for zero dose, heteroscedastic linear regression for log dose, multinomial regression for patients who recently received very low or high doses; and (D) ordinal logistic regression.

RESULTS

For this dataset, method (C) was the only approach that provided a robust estimate of the mortality hazard ratio (HR), with less-extreme weights in a fully weighted analysis and no substantial change of the HR point estimate after weight truncation. After truncating IP weights at the 95th percentile, estimates were similar across the methods.

CONCLUSIONS

EHR data can be used to emulate target trials estimating the comparative effectiveness of dynamic strategies adjusting treatment to evolving patient characteristics. However, model checking, monitoring of large weights, and adaptation of model strategies to account for these is essential if an aspect of treatment is continuous.

Causal inference in survival analysis using longitudinal observational data: Sequential trials and marginal structural models

Longitudinal observational data on patients can be used to investigate causal effects of time-varying treatments on time-to-event outcomes. Several methods have been developed for estimating such effects by controlling for the time-dependent confounding that typically occurs. The most commonly used is marginal structural models (MSM) estimated using inverse probability of treatment weights (IPTW) (MSM-IPTW). An alternative, the sequential trials approach, is increasingly popular, and involves creating a sequence of "trials" from new time origins and comparing treatment initiators and non-initiators. Individuals are censored when they deviate from their treatment assignment at the start of each "trial" (initiator or noninitiator), which is accounted for using inverse probability of censoring weights. The analysis uses data combined across trials. We show that the sequential trials approach can estimate the parameters of a particular MSM. The causal estimand that we focus on is the marginal risk difference between the sustained treatment strategies of "always treat" vs "never treat." We compare how the sequential trials approach and MSM-IPTW estimate this estimand, and discuss their assumptions and how data are used differently. The performance of the two approaches is compared in a simulation study. The sequential trials approach, which tends to involve less extreme weights than MSM-IPTW, results in greater efficiency for estimating the marginal risk difference at most follow-up times, but this can, in certain scenarios, be reversed at later time points and relies on modelling assumptions. We apply the methods to longitudinal observational data from the UK Cystic Fibrosis Registry to estimate the effect of dornase alfa on survival.

Variability of the response to immunotherapy among subgroups of patients with multiple sclerosis

BACKGROUND AND PURPOSE

This study assessed the effect of patient characteristics on the response to disease-modifying therapy (DMT) in multiple sclerosis (MS).

METHODS

We extracted data from 61,810 patients from 135 centers across 35 countries from the MSBase registry. The selection criteria were: clinically isolated syndrome or definite MS, follow-up ≥ 1 year, and Expanded Disability Status Scale (EDSS) score ≥ 3, with ≥1 score recorded per year. Marginal structural models with interaction terms were used to compare the hazards of 12-month confirmed worsening and improvement of disability, and the incidence of relapses between treated and untreated patients stratified by their characteristics.

RESULTS

Among 24,344 patients with relapsing MS, those on DMTs experienced 48% reduction in relapse incidence (hazard ratio [HR] = 0.52, 95% confidence interval [CI] = 0.45-0.60), 46% lower risk of disability worsening (HR = 0.54, 95% CI = 0.41-0.71), and 32% greater chance of disability improvement (HR = 1.32, 95% CI = 1.09-1.59). The effect of DMTs on EDSS worsening and improvement and the risk of relapses was attenuated with more severe disability. The magnitude of the effect of DMT on suppressing relapses declined with higher prior relapse rate and prior cerebral magnetic resonance imaging activity. We did not find any evidence for the effect of age on the effectiveness of DMT. After inclusion of 1985 participants with progressive MS, the effect of DMT on disability mostly depended on MS phenotype, whereas its effect on relapses was driven mainly by prior relapse activity.

CONCLUSIONS

DMT is generally most effective among patients with lower disability and in relapsing MS phenotypes. There is no evidence of attenuation of the effect of DMT with age.

Consistent inverse probability of treatment weighted estimation of the average treatment effect with mismeasured time-dependent confounders

In longitudinal studies, the inverse probability of treatment weighted (IPTW) method is commonly employed to estimate the effect of time-dependent treatments on an outcome of interest. However, it has been documented that when the confounders are subject to measurement error, the naive IPTW method which simply ignores measurement error leads to biased treatment effect estimation. In the existing literature, there is a lack of measurement error correction methods that fully remove measurement error effect and produce consistent treatment effect estimation. In this article, we develop a novel consistent IPTW estimation procedure for longitudinal studies. The key step of the proposed method is to use the observed data to construct a corrected function that is unbiased of the unknown IPTW function. Simulation studies reveal that the proposed method outperforms the existing consistent and approximate measurement error correction methods for IPTW estimation of the average treatment effect. Finally, we apply the proposed method to analyze a real dataset.

Adherence to a priori-Defined Diet Quality Indices Throughout the Early Disease Course Is Associated With Survival in Head and Neck Cancer Survivors: An Application Involving Marginal Structural Models

No studies, to date, have scrutinized the role of a priori dietary patterns on prognosis following a head and neck squamous cell carcinoma (HNSCC) diagnosis. The purpose of this analysis was to evaluate the associations between adherence to six a priori defined diet quality indices (including AHEI-2010, aMED, DASH, and three low-carbohydrate indices) throughout the first 3 years of observation and all-cause and cancer-specific mortalities in 468 newly diagnosed HNSCC patients from the University of Michigan Head and Neck Specialized Program of Research Excellence (UM-SPORE). The dietary intake data were measured using a food frequency questionnaire administered at three annual time points commencing at study entry. Deaths and their causes were documented throughout the study using various data sources. Marginal structural Cox proportional hazards models were used to evaluate the role of diet quality, as a time-varying covariate, on mortality. There were 93 deaths from all causes and 74 cancer-related deaths adjudicated throughout the observation period. There was a strong inverse association between adherence to the AHEI-2010, all-cause mortality (HR_Q5–Q1:0.07, 95% CI:0.01–0.43, p_trend:0.04), and cancer-specific mortality (HR_Q5–Q1:0.15, 95% CI:0.02–1.07, p_trend:0.04). Other more modest associations were noted for the low-carbohydrate indices. In sum, higher adherence to the AHEI-2010 and a plant-based low-carbohydrate index throughout the first 3 years since diagnosis may bolster survival and prognosis in newly diagnosed patients with HNSCC.

Association of Statins and 28-Day Mortality Rates in Patients Hospitalized With Severe Acute Respiratory Syndrome Coronavirus 2 Infection

BACKGROUND

Statins may be protective in severe acute respiratory syndrome coronavirus 2 SARS-CoV-2 infection. The aim of the current study was to evaluate the effect of in-hospital statin use on 28-day mortality rates and intensive care unit (ICU) admission among patients with SARS-CoV-2, stratified into 4 groups: those who used statins before hospitalization (treatment continued or discontinued in the hospital) and those who did not (treatment newly initiated in the hospital or never initiated).

METHODS

In a cohort study of 1179 patients with SARS-CoV-2, record review was used to assess demographics, laboratory measurements, comorbid conditions, and time from admission to death, ICU admission, or discharge. Using marginal structural Cox models, we estimated hazard ratios (HRs) for death and ICU admission.

RESULTS

Among 1179 patients, 676 (57%) were male, 443 (37%) were >65 years old, and 493 (46%) had a body mass index ≥30 (calculated as weight in kilograms divided by height in meters squared). Inpatient statin use reduced the hazard of death (HR, 0.566; P=.008). This association held among patients who did and those who did not use statins before hospitalization (HR, 0.270 [P=.003] and 0.493 [P=.04], respectively). Statin use was associated with improved time to death for patients aged >65 years but not for those ≤65 years old.

CONCLUSION

Statin use during hospitalization for SARS-CoV-2 infection was associated with reduced 28-day mortality rates. Well-designed randomized control trials are needed to better define this relationship.

Sunscreens With High Versus Low Sun Protection Factor and Cutaneous Squamous Cell Carcinoma Risk: A Population-Based Cohort Study

Evidence on sunscreen use and cutaneous squamous cell carcinoma (cSCC) risk is limited. Most studies have not taken sun protection factor (SPF) into consideration and used nonusers of sunscreen as the reference group. Nonusers are likely a priori at lower cSCC risk than users. No study has investigated the effect of high- versus low-SPF sunscreens on cSCC, appropriately adjusting for time-varying confounding. Using data from the Norwegian Women and Cancer Study (1991–2016), we investigated whether use of SPF ≥15 versus SPF <15 sunscreens reduces cSCC risk. We used a marginal structural Cox proportional hazards model with inverse probability of treatment and censoring weights to estimate hazard ratios (HRs) and 95% confidence intervals (CIs). During follow-up of 148,781 women (mean follow-up, 14.3 years), 653 women were diagnosed with cSCC. The effect on cSCC risk of sunscreens with SPF ≥15 versus SPF <15 was close to the null when used at any latitudes (HR = 1.02, 95% CI: 0.82, 1.27) and when used in lower-latitude settings (HR = 1.05, 95% CI: 0.84, 1.32). In conclusion, we found no indication that sunscreens with SPF ≥15 reduced Norwegian women’s cSCC risk more than sunscreens with SPF <15, suggesting that either there is no difference in their effects long-term or the difference is diluted by incorrect application.

Modeling treatment effect modification in multidrug-resistant tuberculosis in an individual patientdata meta-analysis

Effect modification occurs while the effect of the treatment is not homogeneous across the different strata of patient characteristics. When the effect of treatment may vary from individual to individual, precision medicine can be improved by identifying patient covariates to estimate the size and direction of the effect at the individual level. However, this task is statistically challenging and typically requires large amounts of data. Investigators may be interested in using the individual patient data from multiple studies to estimate these treatment effect models. Our data arise from a systematic review of observational studies contrasting different treatments for multidrug-resistant tuberculosis, where multiple antimicrobial agents are taken concurrently to cure the infection. We propose a marginal structural model for effect modification by different patient characteristics and co-medications in a meta-analysis of observational individual patient data. We develop, evaluate, and apply a targeted maximum likelihood estimator for the doubly robust estimation of the parameters of the proposed marginal structural model in this context. In particular, we allow for differential availability of treatments across studies, measured confounding within and across studies, and random effects by study.

Simulating longitudinal data from marginal structural models using the additive hazard model

Observational longitudinal data on treatments and covariates are increasingly used to investigate treatment effects, but are often subject to time-dependent confounding. Marginal structural models (MSMs), estimated using inverse probability of treatment weighting or the g-formula, are popular for handling this problem. With increasing development of advanced causal inference methods, it is important to be able to assess their performance in different scenarios to guide their application. Simulation studies are a key tool for this, but their use to evaluate causal inference methods has been limited. This paper focuses on the use of simulations for evaluations involving MSMs in studies with a time-to-event outcome. In a simulation, it is important to be able to generate the data in such a way that the correct forms of any models to be fitted to those data are known. However, this is not straightforward in the longitudinal setting because it is natural for data to be generated in a sequential conditional manner, whereas MSMs involve fitting marginal rather than conditional hazard models. We provide general results that enable the form of the correctly specified MSM to be derived based on a conditional data generating procedure, and show how the results can be applied when the conditional hazard model is an Aalen additive hazard or Cox model. Using conditional additive hazard models is advantageous because they imply additive MSMs that can be fitted using standard software. We describe and illustrate a simulation algorithm. Our results will help researchers to effectively evaluate causal inference methods via simulation.

Association of Time-Updated Anion Gap With Risk of Kidney Failure in Advanced CKD: A Cohort Study

RATIONALE AND OBJECTIVE

High anion gap acidosis frequently develops in patients with advanced chronic kidney disease (CKD) and might be involved in kidney injury. Its impact on kidney outcomes, however, has not been well studied. We sought to examine the association between time-updated anion gap and the risk of kidney failure with replacement therapy (KFRT) among patients with advanced CKD.

STUDY DESIGN

Retrospective cohort study.

SETTING AND PARTICIPANTS

1,168 patients with CKD stages G3b-G5 who had available data on anion gap.

EXPOSURE

High time-updated anion gap defined as values ≥9.2 (top 25th percentile).

OUTCOMES

KFRT and death.

ANALYTICAL APPROACH

Marginal structural models (MSM) were fit to characterize the association between anion gap and study outcomes while accounting for potential time-dependent confounding.

RESULTS

The mean baseline eGFR of the study participants was 28 mL/min/1.73m². Over a median follow-up of 3.1 years, 317 patients progressed to KFRT (7.5/100 patient-years) and 146 died (3.5/100 patient-years). In the MSM, a high anion gap was associated with a higher rate of KFRT (hazard ratio [HR], 3.04; 95% confidence interval [CI], 1.94-4.75; P<0.001). This association was stronger in patients with baseline eGFR of <30 mL/min/1.73m² (P for interaction=0.05). High anion gap was also associated with a higher mortality rate (HR, 5.56; 95% CI, 2.95-10.5; P<0.001). Sensitivity analyses with different definitions of high anion gap showed similar results.

LIMITATIONS

Observational study design; selection bias due clinical indications for measuring anion gap.

CONCLUSION

Among patients with advanced CKD, high anion gap was associated with an increased risk of progression to KFRT and death.

Using generalized linear models to implement g-estimation for survival data with time-varying confounding

Using data from observational studies to estimate the causal effect of a time-varying exposure, repeatedly measured over time, on an outcome of interest requires careful adjustment for confounding. Standard regression adjustment for observed time-varying confounders is unsuitable, as it can eliminate part of the causal effect and induce bias. Inverse probability weighting, g-computation, and g-estimation have been proposed as being more suitable methods. G-estimation has some advantages over the other two methods, but until recently there has been a lack of flexible g-estimation methods for a survival time outcome. The recently proposed Structural Nested Cumulative Survival Time Model (SNCSTM) is such a method. Efficient estimation of the parameters of this model required bespoke software. In this article we show how the SNCSTM can be fitted efficiently via g-estimation using standard software for fitting generalised linear models.The ability to implement g-estimation for a survival outcome using standard statistical software greatly increases the potential uptake of this method. We illustrate the use of this method of fitting the SNCSTM by reanalyzing data from the UK Cystic Fibrosis Registry, and provide example R code to facilitate the use of this approach by other researchers.

Dealing With Treatment-Confounder Feedback and Sparse Follow-up in Longitudinal Studies: Application of a Marginal Structural Model in a Multiple Sclerosis Cohort

The beta-interferons are widely prescribed platform therapies for patients with multiple sclerosis (MS). We accessed a cohort of patients with relapsing-onset MS from British Columbia, Canada (1995-2013), to examine the potential survival advantage associated with beta-interferon exposure using a marginal structural model. Accounting for potential treatment-confounder feedback between comorbidity, MS disease progression, and beta-interferon exposure, we found an association between beta-interferon exposure of at least 6 contiguous months and improved survival (hazard ratio (HR) = 0.63, 95% confidence interval 0.47, 0.86). We also assessed potential effect modifications by sex, baseline age, or baseline disease duration, and found these factors to be important effect modifiers. Sparse follow-up due to variability in patient contact with the health system is one of the biggest challenges in longitudinal analyses. We considered several single-level and multilevel multiple imputation approaches to deal with sparse follow-up and disease progression information; both types of approach produced similar estimates. Compared to ad hoc imputation approaches, such as linear interpolation (HR = 0.63), and last observation carried forward (HR = 0.65), all multiple imputation approaches produced a smaller hazard ratio (HR = 0.53), although the direction of effect and conclusions drawn concerning the survival advantage remained the same.

Risk of Tumor Onset in HIV+ Patients on Two-Drug Regimens: A Cohort Study in an Italian Hospital

Currently approved 2-drug therapies are as effective as 3-drug regimens but could potentially lead to increased cancer risk due to less efficient immune recovery. We conducted a longitudinal cohort study in a tertiary Italian hospital to investigate HIV+ patients starting a triple therapy (TT) (2 NRTIs +3rd agent) or a dual therapy (DT) (3TC/FTC+boosted-PI, boosted-DRV+RAL, and 3TC/FTC or RPV+DTG) regimen between 2009 and 2018. The effect of DT (vs. TT) on tumor onset was evaluated by the multivariable Cox regression and the marginal structural Cox model, after estimating the inverse probability of treatment weights (IPTW). One thousand one hundred and seven patients who had a median follow-up of 4.2 person-years (py) were evaluated; 69.2% were males, with a median age of 43 years. Overall 2,513 treatments were started during the study period (479 DT, 2,034 TT). Eight tumors occurred over 965 py with DT and 35 over 3,817 py during TT (p = .797). In the Cox regression, DT did not predict an increased risk of tumor compared with TT (HR 1.14; p = .757) after adjusting for potential confounders. A marginal structural model using IPTW (HR 0.68; p = .328) and stabilized IPTW (HR 0.69; p = .361) confirmed this result. Preliminary findings from our cohort do not suggest an increased risk of tumors with DT compared to TT.

Variable selection for causal mediation analysis using LASSO-based methods

Causal mediation effect estimates can be obtained from marginal structural models using inverse probability weighting with appropriate weights. In order to compute weights, treatment and mediator propensity score models need to be fitted first. If the covariates are high-dimensional, parsimonious propensity score models can be developed by regularization methods including LASSO and its variants. Furthermore, in a mediation setup, more efficient direct or indirect effect estimators can be obtained by using outcome-adaptive LASSO to select variables for propensity score models by incorporating the outcome information. A simulation study is conducted to assess how different regularization methods can affect the performance of estimated natural direct and indirect effect odds ratios. Our simulation results show that regularizing propensity score models by outcome-adaptive LASSO can improve the efficiency of the natural effect estimators and by optimizing balance in the covariates, bias can be reduced in most cases. The regularization methods are then applied to MIMIC-III database, an ICU database developed by MIT.

Is soft drink consumption associated with gestational hypertension? Results from the BRISA cohort

It is still unknown whether excessive consumption of sugar-sweetened beverages may be linked to gestational hypertensive disorders, other than preeclampsia. This study investigated the association between soft drink consumption and hypertension during pregnancy, analyzing the relationship from the perspective of counterfactual causal theory. Data from pregnant women of the BRISA cohort were analyzed (1,380 in São Luis and 1,370 in Ribeirão Preto, Brazil). The explanatory variable was the frequency of soft drink consumption during pregnancy obtained in a prenatal interview. The outcome was gestational hypertension based on medical diagnosis, at the time of delivery. A theoretical model of the association between soft drink consumption and gestational hypertension was constructed using a directed acyclic graph. Marginal structural models (MSM) weighted by the inverse of the probability of soft drink consumption were also employed. Using Poisson regression analysis, high soft drink consumption (≥7 times/week) was associated with gestational hypertension in São Luís (RR=1.48; 95%CI: 1.03-2.10), in Ribeirão Preto (RR=1.51; 95%CI: 1.13-2.01), and in the two cohorts combined (RR=1.45; 95%CI: 1.16-1.82) compared to lower exposure (<7 times/week). In the MSM, the association between high soft drink consumption and gestational hypertension was observed in Ribeirão Preto (RR=1.63; 95%CI: 1.21-2.19) and in the two cohorts combined (RR=1.51; 95%CI: 1.15-1.97), but not in São Luís (RR=1.26; 95%CI: 0.79-2.00). High soft drink consumption seems to be a risk factor for gestational hypertension, suggesting that it should be discouraged during pregnancy.

Poverty increases the risk of incident cognitive impairment among older adults: a longitudinal study in China

Objectives: With a longitudinal design, we aimed to investigate the relationship between poverty and the risk of incident cognitive impairment in China.Methods: We used three waves of the Chinese Longitudinal Healthy Longevity Survey (2008-2014). Cognitive impairment was assessed using the Chinese version of the Mini Mental State Examination. Poverty was measured according to the latest national poverty line settled at an annual per-capita income of 2300 yuan (approximately equivalent to 1.25 dollar/day) in 2011 in China. A marginal structural model was utilized to explore the association between poverty and the risk of incident cognitive impairment. The subgroup analyses were also conducted in this study.Results: The cumulative incidence of cognitive impairment over 6 years was 30.69% (1936/6309). Poverty increased 34% risk of incident cognitive impairment in the elderly (odds ratio = 1.34, 95% confidence interval (CI): 1.15-1.56) after controlling behavioral factors and health status covariates. Participants who were male (OR = 1.38, 95% CI: 1.08-1.76), lived in urban areas (OR = 1.55, 95% CI: 1.22-1.98), and were married (OR = 1.72, 95% CI: 1.28-2.32) had higher poverty risks on incident cognitive impairment in subgroup analyses.Conclusions: Our results provide empirical support for the ongoing discussion about how economic hardship impacts of cognitive functioning, and highlight the negative health risks that economically disadvantaged individuals may experience.

Investigating the Impact of Early Valve Surgery on Survival in Staphylococcus aureus Infective Endocarditis Using a Marginal Structural Model Approach: Results of a Large, Prospectively Evaluated Cohort

BACKGROUND

The impact of valve surgery on outcomes of Staphylococcus aureus infective endocarditis (SAIE) remains controversial. We tested the hypothesis that early valve surgery (EVS) improves survival by using a novel approach that allows for inclusion of major confounders in a time-dependent way.

METHODS

EVS was defined as valve surgery within 60 days. Univariable and multivariable Cox regression analyses were performed. To account for treatment selection bias, we additionally used a weighted Cox model (marginal structural model) that accounts for time-dynamic imbalances between treatment groups. To address survivor bias, EVS was included as a time-dependent variable. Follow-up of patients was 1 year.

RESULTS

Two hundred and three patients were included in the analysis; 50 underwent EVS. All-cause mortality at day 30 was 26%. In the conventional multivariable Cox regression model, the effect of EVS on the death hazard was 0.85 (95% confidence interval [CI], .47-1.52). Using the weighted Cox model, the death hazard rate (HR) of EVS was 0.71 (95% CI, .34-1.49). In subgroup analyses, no survival benefit was observed in patients with septic shock (HR, 0.80 [CI, .26-2.46]), in NVIE (HR, 0.76 [CI, .33-1.71]) or PVIE (HR, 1.02 [CI, .29-3.54]), or in patients with EVS within 14 days (HR, 0.97 [CI, .46-2.07]).

CONCLUSIONS

Using both a conventional Cox regression model and a weighted Cox model, we did not find a survival benefit for patients who underwent EVS in our cohort. Until results of randomized controlled trials are available, EVS in SAIE should be based on individualized decisions of an experienced multidisciplinary team.

CLINICAL TRIALS REGISTRATION

German Clinical Trials registry (DRKS00005045).

Understanding Marginal Structural Models for Time-Varying Exposures: Pitfalls and Tips

Epidemiologists are increasingly encountering complex longitudinal data, in which exposures and their confounders vary during follow-up. When a prior exposure affects the confounders of the subsequent exposures, estimating the effects of the time-varying exposures requires special statistical techniques, possibly with structural (ie, counterfactual) models for targeted effects, even if all confounders are accurately measured. Among the methods used to estimate such effects, which can be cast as a marginal structural model in a straightforward way, one popular approach is inverse probability weighting. Despite the seemingly intuitive theory and easy-to-implement software, misunderstandings (or "pitfalls") remain. For example, one may mistakenly equate marginal structural models with inverse probability weighting, failing to distinguish a marginal structural model encoding the causal parameters of interest from a nuisance model for exposure probability, and thereby failing to separate the problems of variable selection and model specification for these distinct models. Assuming the causal parameters of interest are identified given the study design and measurements, we provide a step-by-step illustration of generalized computation of standardization (called the g-formula) and inverse probability weighting, as well as the specification of marginal structural models, particularly for time-varying exposures. We use a novel hypothetical example, which allows us access to typically hidden potential outcomes. This illustration provides steppingstones (or "tips") to understand more concretely the estimation of the effects of complex time-varying exposures.

Assessing trends in vaccine efficacy by pathogen genetic distance

Preventive vaccines are an effective public health intervention for reducing the burden of infectious diseases, but have yet to be developed for several major infectious diseases. Vaccine sieve analysis studies whether and how the efficacy of a vaccine varies with the genetics of the infectious pathogen, which may help guide future vaccine development and deployment. A standard statistical approach to sieve analysis compares the effect of the vaccine to prevent infection and disease caused by pathogen types defined dichotomously as genetically near or far from a reference pathogen strain inside the vaccine construct. For example, near may be defined by amino acid identity at all amino acid positions considered in a multiple alignment and far defined by at least one amino acid difference. An alternative approach is to study the efficacy of the vaccine as a function of genetic distance from a pathogen to a reference vaccine strain where the distance cumulates over the set of amino acid positions. We propose a nonparametric method for estimating and testing the trend in the effect of a vaccine across genetic distance. We illustrate the operating characteristics of the estimator via simulation and apply the method to a recent preventive malaria vaccine efficacy trial.

Adjusting for time-varying confounders in survival analysis using structural nested cumulative survival time models

Accounting for time-varying confounding when assessing the causal effects of time-varying exposures on survival time is challenging. Standard survival methods that incorporate time-varying confounders as covariates generally yield biased effect estimates. Estimators using weighting by inverse probability of exposure can be unstable when confounders are highly predictive of exposure or the exposure is continuous. Structural nested accelerated failure time models (AFTMs) require artificial recensoring, which can cause estimation difficulties. Here, we introduce the structural nested cumulative survival time model (SNCSTM). This model assumes that intervening to set exposure at time t to zero has an additive effect on the subsequent conditional hazard given exposure and confounder histories when all subsequent exposures have already been set to zero. We show how to fit it using standard software for generalized linear models and describe two more efficient, double robust, closed-form estimators. All three estimators avoid the artificial recensoring of AFTMs and the instability of estimators that use weighting by the inverse probability of exposure. We examine the performance of our estimators using a simulation study and illustrate their use on data from the UK Cystic Fibrosis Registry. The SNCSTM is compared with a recently proposed structural nested cumulative failure time model, and several advantages of the former are identified.

Cognitive reserve regulates the association between hearing difficulties and incident cognitive impairment evidence from a longitudinal study in China

BACKGROUND

Cognitive reserve (CR) can prevent the risk of incident cognitive impairment in the elderly. However, the moderator effects of CR on the link between hearing difficulties (HDs) and the risk of incident cognitive impairment are not well understood.

METHODS

This cohort study obtained data from the Chinese Longitudinal Healthy Longevity Survey from 2008 to 2014. The baseline samples included 6309 participants aged 65 years and older at baseline. Cognitive impairment was assessed using the Chinese version of the Mini Mental State Examination. A composite measure of CR was calculated based on education, occupational complexity, and leisure activities of the participants. The marginal structural model was utilized to investigate whether CR moderates the association between HD and incident cognitive impairment. Odds ratios (ORs) and accompanying 95% confidence intervals (CIs) were calculated.

RESULTS

Of the 6309 participants at baseline, 1936 (30.7%) developed cognitive impairment during the 6-year follow-up period and 2562 (40.6%) reported HD. The risk of incident cognitive impairment was 1.90-fold (95% CI 1.69-2.14) for participants developing HD compared to those without. Those with middle CR had lower OR (0.72, 95% CI 0.62-0.82) that further decreased to 0.58 (95% CI 0.49-0.69) for those with high CR. Participants with HD with low CR showed the highest OR (4.32, 95% CI 3.42-5.47). In addition, individuals with HD with low education levels or low complex occupations had the highest risk of incident cognitive impairment.

CONCLUSIONS

CR moderates the negative association between HD and cognitive function. Education and occupation complexity may be more sensitive proxies for CR.

Examining the trimester-specific effects of low gestational weight gain on birthweight: the BOSHI study

Low gestational weight gain (GWG) is a known risk factor of low birthweight. Although studies have previously examined the associations between GWG and birthweight, the period-specific effects of low GWG in each trimester remain unclear. This study aimed to quantify the trimester-specific direct effects of low GWG in Japanese women on birthweight. Using perinatal data from a cohort study, we analyzed pregnant women delivered at an obstetrics/gynecology hospital between October 2006 and May 2010. We focused on women with a pre-pregnancy body mass index (BMI) below 25 kg/m2. The exposure was low GWG. The gestation period was subdivided into trimesters, and the direct effects of low trimester-specific GWG on birthweight were estimated using marginal structural models. These models were guided by a direct acyclic graph that incorporated potential confounders, including pre-pregnancy BMI, age, smoking during pregnancy, height, and parity. We analyzed 563 women and their families. The mean cumulative GWG by the end of the first, second, and third trimesters was 0.9, 6.2, and 10.7 kg, respectively. Approximately 14.0% of the women gained total weight below the range recommended by Japanese Ministry of Health, Labour and Welfare. The direct effects of low GWG on birthweight were 65.9 g (95% confidence interval: 11.4, 120.5), -195.4 g (-263.4, -127.4), and -188.8 g (-292.0, -85.5) for the first, second, and third trimesters, respectively. Insufficient weight gain in the second and third trimesters had a negative impact on birthweight after adjusting for pre-pregnancy BMI and other covariates.

Impact of Cumulative Unhealthy Sleep Practices in Adolescence on Substance Use in Young Adulthood Estimated Using Marginal Structural Modeling

Objectives

The purpose of this study was to identify the impact of chronic, unhealthy sleep practices in adolescence on substance use in young adulthood. Unhealthy sleep practices in adolescent samples exhibit a bidirectional relationship with substance use. The relationship is further complicated if we consider that confounders such as depression vary over time and are often in response to adolescents’ prior poor sleep practice, which can be addressed by a counterfactual approach using a marginal structural model.

Methods

Data in this study are from the Taiwan Youth Project, a longitudinal study that started in 2000 and surveyed 2,690 7th grade students at age 13. Outcomes include frequency of cigarette smoking and alcohol drinking at age 21. Three unhealthy sleep practices were included in this study: short sleep, social jetlag, and sleep disturbance. We used a marginal structural model with stabilized inverse probability-of-treatment weights to address time-varying confounders in each wave and a total sample of 1,678 adolescents with complete information for this study.

Results

Accumulated waves of sleep disturbance and social jetlag in adolescence were significantly associated with cigarette use in young adulthood. Accumulated social jetlag but not sleep disturbance was also associated with alcohol use in adulthood. Accumulated waves of short sleep were not associated with later alcohol use, but were negatively correlated with cigarette use.

Conclusion

Interventions that aim to reduce the likelihood of substance use in young adulthood should consider confronting unhealthy sleep practices, in particular the discrepancy between bedtimes on school days and weekends and sleep disturbance.

Comparing two counterfactual-outcome approaches in causal mediation analysis of a multicategorical exposure: An application for the estimation of the effect of maternal intake of inhaled corticosteroids doses on birthweight

Although medical research frequently involves an exposure variable with three or more discrete levels, detailed presentations of mediation techniques for dealing with multicategorical (multilevel) exposures are sparse. In this paper, we study two causal mediation approaches applicable to such a type of exposure for continuous mediator and outcome: the closed-form regression-based approach of Valeri and VanderWeele, and the marginal structural model-based approach of Lange, Vansteelandt, and Bekaert. While the consideration of multicategorical exposures is found explicitly addressed in the literature for the latter approach, this is, to our knowledge, not yet the case for the former. We first illustrate the application of the two aforementioned approaches to assess the dose-response relationship between maternal intake of inhaled corticosteroids and birthweight, where this relationship is potentially mediated by gestational age. More specifically, we provide a precise roadmap for the application of the regression-based approach and of the marginal structural model-based approach on our cohort of pregnancies. Expressions for the natural direct and indirect effects associated with our categorical exposure are provided and, for the regression-based approach, analytic formulas for standard error calculation using the delta method are presented for these effects. Second, a simulation study which mimics our data is presented to add to current knowledge on these causal mediation techniques. Results from this study highlight the relevance to assess robustness of mediation results obtained from multicategorical exposures, most notably for the least prevalent of exposure categories.

Parental Obesity and Offspring Pubertal Development: Project Viva

OBJECTIVE

To investigate the association of preconception parental obesity (body mass index [BMI] ≥30 kg/m²) with offspring pubertal development.

STUDY DESIGN

Among 1377 children from a prospective prebirth cohort in Boston, we examined markers of puberty (age at peak height velocity [PHV], age at menarche, self-reported pubertal development score), and adrenarche (pictograph Tanner pubic hair staging). We used multivariable regression models to examine associations of maternal and paternal obesity with offspring pubertal indices, and applied marginal structural models to estimate the controlled direct effect not mediated by offspring prepubertal BMI.

RESULTS

The prevalence of paternal obesity alone, maternal obesity alone, and biparental obesity were 10.5%, 10.1%, and 5%, respectively. After adjusting for demographic and socioeconomic factors, parental heights and maternal age at menarche, maternal obesity alone (vs neither parent with obesity) was associated with earlier age at PHV (β -0.30 years; 95% CI -0.57, -0.03) and higher early adolescent pubertal score (0.29 units; 0.10, 0.48) in boys, but not with pubertal or adrenarchal outcomes in girls. Paternal obesity alone was not associated with any outcomes in either boys or girls. Biparental obesity was associated with earlier age at PHV in boys and earlier menarche in girls. Using marginal structural models with stabilized inverse probability weighting, maternal obesity alone had significant controlled direct effects on age at PHV (-0.31 years; -0.62, 0.00) and on pubertal score (0.22 units; 0.00, 0.44) in boys, independent of prepubertal BMI.

CONCLUSION

Maternal, but not paternal, obesity is associated with earlier pubertal development in boys, and such association is independent of prepubertal BMI.

Causal comparative effectiveness analysis of dynamic continuous-time treatment initiation rules with sparsely measured outcomes and death

Evidence supporting the current World Health Organization recommendations of early antiretroviral therapy (ART) initiation for adolescents is inconclusive. We leverage a large observational data and compare, in terms of mortality and CD4 cell count, the dynamic treatment initiation rules for human immunodeficiency virus-infected adolescents. Our approaches extend the marginal structural model for estimating outcome distributions under dynamic treatment regimes, developed in Robins et al. (2008), to allow the causal comparisons of both specific regimes and regimes along a continuum. Furthermore, we propose strategies to address three challenges posed by the complex data set: continuous-time measurement of the treatment initiation process; sparse measurement of longitudinal outcomes of interest, leading to incomplete data; and censoring due to dropout and death. We derive a weighting strategy for continuous-time treatment initiation, use imputation to deal with missingness caused by sparse measurements and dropout, and define a composite outcome that incorporates both death and CD4 count as a basis for comparing treatment regimes. Our analysis suggests that immediate ART initiation leads to lower mortality and higher median values of the composite outcome, relative to other initiation rules.

Effortless or less effort? Effects of tracks on students' engagement

BACKGROUND

It is found that assigning students to a lower track during secondary education negatively affects their academic performance. As an explanation, it is often mentioned that an anti-school culture in lower tracks undermines students' effort and involvement.

AIMS

This study assessed whether going to a lower track affects student engagement. For if an anti-school culture is to blame for limiting lower track students' performance, lower track assignment should reduce engagement.

SAMPLE

A sample of a longitudinal cohort study during secondary education in Flanders (northern Belgium) was used to describe development in engagement with n = 5,417 students in 46 schools. Four tracks were investigated across four school years.

METHOD

Two main methodological challenges were present in this study, different student intake in each track and many students changing from a higher to lower track over time. Accordingly, we used inverse probability treatment weights with marginal structural mean models to account for different student intake and track changes. A comparison was made per pair of tracks that are hierarchically consecutive by matching students who were comparable across these tracks. Accordingly, there were three pairwise comparisons.

RESULTS

It was never found that being continuously in lower track negatively affects engagement. Only for one pairwise comparison, there was evidence that students who changed from the higher to lower track had lower engagement.

CONCLUSIONS

We rejected the hypothesis that lower tracks negatively affect student engagement. This makes the anti-school culture as an explanation for lower track assignment negatively affecting academic performance implausible.

Increased non-AIDS mortality among persons with AIDS-defining events after antiretroviral therapy initiation

INTRODUCTION

HIV-1 infection leads to chronic inflammation and to an increased risk of non-AIDS mortality. Our objective was to determine whether AIDS-defining events (ADEs) were associated with increased overall and cause-specific non-AIDS related mortality after antiretroviral therapy (ART) initiation.

METHODS

We included HIV treatment-naïve adults from the Antiretroviral Therapy Cohort Collaboration (ART-CC) who initiated ART from 1996 to 2014. Causes of death were assigned using the Coding Causes of Death in HIV (CoDe) protocol. The adjusted hazard ratio (aHR) for overall and cause-specific non-AIDS mortality among those with an ADE (all ADEs, tuberculosis (TB), Pneumocystis jiroveci pneumonia (PJP), and non-Hodgkin's lymphoma (NHL)) compared to those without an ADE was estimated using a marginal structural model.

RESULTS

The adjusted hazard of overall non-AIDS mortality was higher among those with any ADE compared to those without any ADE (aHR 2.21, 95% confidence interval (CI) 2.00 to 2.43). The adjusted hazard of each of the cause-specific non-AIDS related deaths were higher among those with any ADE compared to those without, except metabolic deaths (malignancy aHR 2.59 (95% CI 2.13 to 3.14), accident/suicide/overdose aHR 1.37 (95% CI 1.05 to 1.79), cardiovascular aHR 1.95 (95% CI 1.54 to 2.48), infection aHR (95% CI 1.68 to 2.81), hepatic aHR 2.09 (95% CI 1.61 to 2.72), respiratory aHR 4.28 (95% CI 2.67 to 6.88), renal aHR 5.81 (95% CI 2.69 to 12.56) and central nervous aHR 1.53 (95% CI 1.18 to 5.44)). The risk of overall and cause-specific non-AIDS mortality differed depending on the specific ADE of interest (TB, PJP, NHL).

CONCLUSIONS

In this large multi-centre cohort collaboration with standardized assignment of causes of death, non-AIDS mortality was twice as high among patients with an ADE compared to without an ADE. However, non-AIDS related mortality after an ADE depended on the ADE of interest. Although there may be unmeasured confounders, these findings suggest that a common pathway may be independently driving both ADEs and NADE mortality. While prevention of ADEs may reduce subsequent death due to NADEs following ART initiation, modification of risk factors for NADE mortality remains important after ADE survival.

Emulating a trial of joint dynamic strategies: An application to monitoring and treatment of HIV-positive individuals

Decisions about when to start or switch a therapy often depend on the frequency with which individuals are monitored or tested. For example, the optimal time to switch antiretroviral therapy depends on the frequency with which HIV-positive individuals have HIV RNA measured. This paper describes an approach to use observational data for the comparison of joint monitoring and treatment strategies and applies the method to a clinically relevant question in HIV research: when can monitoring frequency be decreased and when should individuals switch from a first-line treatment regimen to a new regimen? We outline the target trial that would compare the dynamic strategies of interest and then describe how to emulate it using data from HIV-positive individuals included in the HIV-CAUSAL Collaboration and the Centers for AIDS Research Network of Integrated Clinical Systems. When, as in our example, few individuals follow the dynamic strategies of interest over long periods of follow-up, we describe how to leverage an additional assumption: no direct effect of monitoring on the outcome of interest. We compare our results with and without the "no direct effect" assumption. We found little differences on survival and AIDS-free survival between strategies where monitoring frequency was decreased at a CD4 threshold of 350 cells/μl compared with 500 cells/μl and where treatment was switched at an HIV-RNA threshold of 1000 copies/ml compared with 200 copies/ml. The "no direct effect" assumption resulted in efficiency improvements for the risk difference estimates ranging from an 7- to 53-fold increase in the effective sample size.

Hospitalization risk in bipolar disorder patients treated with lurasidone versus other atypical antipsychotics

OBJECTIVE

This observational study compared the risk of hospitalization for patients with bipolar disorder when treated with lurasidone versus other oral atypical antipsychotics.

METHODS

This US commercial claims analysis (4 April 2010 through 24 September 2014) used the Optum Research Database to identify adult patients with bipolar disorder treated with oral atypical antipsychotics (N = 11,132). The first claim for an atypical antipsychotic defined the index date, with pre-index and post-index periods of 180 and 360 days, respectively. Every month of the post-index period was categorized as monotherapy treatment with lurasidone, aripiprazole, olanzapine, quetiapine, risperidone, ziprasidone, no/minimal treatment or other. Starting with the initial month of treatment, the risk of psychiatric or all-cause hospitalization in the subsequent month was examined based on treatment in the current month and pre-index covariates (age, gender, hospitalizations, emergency room visits, diagnoses for anxiety, alcohol abuse, substance abuse, hypertension, type 2 diabetes and obesity) and time-varying versions of the pre-index covariates using a marginal structural model.

RESULTS

After controlling for covariates, relative to lurasidone, the odds of psychiatric and all-cause hospitalization, respectively, were 2-3 times higher for olanzapine (odds ratio [OR] = 2.78, CI 1.09, 7.08, p = .032; OR = 3.20, CI 1.24, 8.26, p = .016), quetiapine (OR = 2.80, CI 1.13, 6.95, p = .026; OR = 3.23, CI 1.29, 8.11, p = .013), risperidone (OR = 2.50, CI 1.01, 6.21, p = .048; OR = 2.79, CI 1.11, 7.02, p = .029), aripiprazole (OR = 2.13, CI 0.87, 5.20, p = .097; OR = 2.57, CI 1.04, 6.37, p = .041) and ziprasidone (OR =2.31, CI 0.91, 5.85, p = .079; OR = 2.49, CI 0.97, 6.40, p = .058).

CONCLUSIONS

In this claims database analysis, lurasidone-treated patients with bipolar disorder had a significantly lower risk of psychiatric hospitalization compared to quetiapine, olanzapine and risperidone, but not aripiprazole or ziprasidone. Lurasidone-treated patients had a significantly lower risk of all-cause hospitalization compared to quetiapine, olanzapine, risperidone and aripiprazole, but not ziprasidone.

Association between depressive symptoms, CD4 count and HIV viral suppression among HIV-HCV co-infected people

Depressive symptoms are associated with poor HIV viral control and immune recovery among people living with HIV. However, no prior studies assessed this association exclusively among people co-infected with HIV-hepatitis C virus (HCV). While people with HIV only and those with HIV-HCV co-infection share many characteristics, co-infected people may become more susceptible to the effects of depressive symptoms on health outcomes. We assessed this association exclusively among people co-infected with HIV-HCV in Canada using data from the Food Security & HIV-HCV Sub-Study (FS Sub-Study) of the Canadian Co-Infection Cohort (CCC). Stabilized inverse probability weighted marginal structural model was used to account for potential time-varying confounders. A total of 725 participants were enrolled between 2012 and 2015. At baseline, 52% of participants reported depressive symptoms, 75% had undetectable HIV viral load, and median CD4 count was 466 (IQR 300-665). People experiencing depressive symptoms had 1.32 times (95% CI: 1.07, 1.63) the risk of having detectable HIV viral load, but had comparable CD4 count to people who did not experience depressive symptoms (fold change of CD4 = 0.96, 95% CI: 0.91, 1.03). Presence of depressive symptoms is a risk factor for incomplete short-term HIV viral suppression among people co-infected with HIV-HCV. Therefore, depressive symptoms screening and related counseling may improve HIV related health outcomes and reduce HIV transmission.

Adherence and virologic outcomes among treatment-naïve veteran patients with human immunodeficiency virus type 1 infection

Many studies have estimated the association between the adherence to antiretroviral therapies and human immunodeficiency virus (HIV) patients' virologic/immunologic outcomes. However, evidence is lacking on the causal effect of adherence on the outcomes. The goal of this study is to understand whether near perfect adherence is necessary to achieve optimal virologic outcome and also to investigate the effect of initial adherence to antiretroviral therapies on initial viral suppression by different regimens. A cohort study was conducted on HIV veterans initiating antiretroviral therapies in 1999 to 2015. The primary outcome was the first viral suppression occurred within 30 to 60 days since the index date. Multiple imputation was used to impute the missing value of virologic outcomes. The inverse probability of treatment weighting (IPTW) method was applied to estimate the viral suppression rate at each specific adherence category for each regimen category. Marginal structural models with IPTW were used to estimate the risk of viral suppression in lower-adherence categories in comparison to near-perfect adherence level ≥95%. Data showed that lower adherence caused lower viral suppression rate, with the association differentiated by the regimen. Patients on integrase strand transfer had the highest viral suppression rate, with patients on protease inhibitors having the lowest rate. Regardless of regimens, the viral suppression rate among patients at initial adherence of 75 to <95% was not statistically different from patients at adherence of ≥95%; however, the differences might be clinically significant.

A simulation study on implementing marginal structural models in an observational study with switching medication based on a biomarker

Assessing treatment effectiveness in longitudinal data can be complex when treatments are not randomly assigned and patients are allowed to switch treatment to other or no treatment, often in a manner that is driven by changes in one or more variables associated with patient or clinical characteristics. There can be confounding of the treatment effect from a time-varying variable, i.e., one which is affected by previous exposure and can in turn also influence subsequent treatment changes. Precision medicine relies on validated biomarkers to better classify patients by their probable response to treatment. However, biomarkers may be a source of time-varying confounding, which are affected by prior treatment in the evaluation and are also subject to measurement errors. The impact of switching medications based on a biomarker has received less attention. We conducted simulation studies to explore biased estimation under various scenarios when marginal structural model estimations are employed. Holding model misspecification issues constant, bias is severe in the presence of multiple switching, along with measurement error and missing data in the covariates.

Maternal gestational weight gain and DNA methylation in young women: application of life course mediation methods

AIM

To investigate the role of maternal gestational weight gain (GWG) and prepregnancy BMI on programming offspring DNA methylation.

METHODS

Among 589 adult (age = 32) women participants of the Jerusalem Perinatal Study, we quantified DNA methylation in five candidate genes. We used inverse probability-weighting and parametric g-formula to estimate direct effects of maternal prepregnancy BMI and GWG on methylation.

RESULTS

Higher maternal GWG, but not prepregnancy BMI, was inversely related to offspring ABCA1 methylation (β = -1.1% per quartile; 95% CI: -2.0, -0.3) after accounting for ancestry, parental and offspring exposures. Total and controlled direct effects were nearly identical suggesting included offspring exposures did not mediate this relationship. Results were robust to sensitivity analyses for missing data and model specification.

CONCLUSION

We find some support for epigenetic programming and highlight strengths and limitations of these methods relative to other prevailing approaches.

Food insecurity may lead to incomplete HIV viral suppression and less immune reconstitution among HIV/hepatitis C virus-coinfected people

OBJECTIVES

The aim of this study was to determine the impact of food insecurity (FI) on HIV viral load and CD4 count among people coinfected with HIV and hepatitis C virus (HCV).

METHODS

This study was conducted using data from the Food Security & HIV-HCV Sub-Study of the Canadian Co-Infection Cohort study. FI was measured using the adult scale of Health Canada's Household Food Security Survey Module and was classified into three categories: food security, moderate food insecurity and severe food insecurity. The association between FI, HIV viral load, and CD4 count was assessed using a stabilized inverse probability weighted marginal structural model.

RESULTS

A total of 725 HIV/HCV-coinfected people with 1973 person-visits over 3 years of follow-up contributed to this study. At baseline, 23% of participants experienced moderate food insecurity and 34% experienced severe food insecurity. The proportion of people with undetectable HIV viral load was 75% and the median CD4 count was 460 [interquartile range (IQR): 300-665] cells/μL. People experiencing severe food insecurity had 1.47 times [95% confidence interval (CI): 1.14, 1.88] the risk of having detectable HIV viral load and a 0.91-fold (95% CI: 0.84, 0.98) increase in CD4 count compared with people who were food secure.

CONCLUSIONS

These findings provide evidence of the negative impact of food insecurity on HIV viral load and CD4 count among HIV/HCV-coinfected people.

Examination of cumulative effects of early adolescent depression on cannabis and alcohol use disorder in late adolescence in a community-based cohort

BACKGROUND AND AIMS

Although they often co-occur, the longitudinal relationship between depression and substance use disorders during adolescence remains unclear. This study estimated the effects of cumulative depression during early adolescence (ages 13-15 years) on the likelihood of cannabis use disorder (CUD) and alcohol use disorder (AUD) at age 18.

DESIGN

Prospective cohort study of youth assessed at least annually between 6th and 9th grades (~ age 12-15) and again at age 18. Marginal structural models based on a counterfactual framework that accounted for both potential fixed and time-varying confounders were used to estimate cumulative effects of depressive symptoms over early adolescence.

SETTING

The sample originated from four public middle schools in Seattle, Washington, USA.

PARTICIPANTS

The sample consisted of 521 youth (48.4% female; 44.5% were non-Hispanic White).

MEASUREMENTS

Structured in-person interviews with youth and their parents were conducted to assess diagnostic symptom counts of depression during early adolescence; diagnoses of CUD and AUD at age 18 was based the Voice-Diagnostic Interview Schedule for Children. Cumulative depression was defined as the sum of depression symptom counts from grades 7-9.

FINDINGS

The past-year prevalence of cannabis and alcohol use disorder at the age 18 study wave was 20.9 and 19.8%, respectively. A 1 standard deviation increase in cumulative depression during early adolescence was associated with a 50% higher likelihood of CUD [prevalence ratio (PR) = 1.50; 95% confidence interval (CI) = 1.07, 2.10]. Although similar in direction, there was no statistically significant association between depression and AUD (PR = 1.41; 95% CI = 0.94, 2.11). Further, there were no differences in associations according to gender.

CONCLUSIONS

Youth with more chronic or severe forms of depression during early adolescence may be at elevated risk for developing cannabis use disorder compared with otherwise similar youth who experience fewer depressive symptoms during early adolescence.

simcausal R Package: Conducting Transparent and Reproducible Simulation Studies of Causal Effect Estimation with Complex Longitudinal Data

The simcausal R package is a tool for specification and simulation of complex longitudinal data structures that are based on non-parametric structural equation models. The package aims to provide a flexible tool for simplifying the conduct of transparent and reproducible simulation studies, with a particular emphasis on the types of data and interventions frequently encountered in real-world causal inference problems, such as, observational data with time-dependent confounding, selection bias, and random monitoring processes. The package interface allows for concise expression of complex functional dependencies between a large number of nodes, where each node may represent a measurement at a specific time point. The package allows for specification and simulation of counterfactual data under various user-specified interventions (e.g., static, dynamic, deterministic, or stochastic). In particular, the interventions may represent exposures to treatment regimens, the occurrence or non-occurrence of right-censoring events, or of clinical monitoring events. Finally, the package enables the computation of a selected set of user-specified features of the distribution of the counterfactual data that represent common causal quantities of interest, such as, treatment-specific means, the average treatment effects and coefficients from working marginal structural models. The applicability of simcausal is demonstrated by replicating the results of two published simulation studies.

Hypertensive Disorders in Pregnancy and the Risk of Subsequent Cardiovascular Disease

BACKGROUND

Hypertensive disorders in pregnancy (HDP) have been shown to predict later risk of cardiovascular disease (CVD). However, previous studies have not accounted for subsequent pregnancies and their complications, which are potential confounders and intermediates of this association.

METHODS

A cohort of 146 748 women with a first pregnancy was constructed using the Clinical Practice Research Datalink. HDP was defined using diagnostic codes, elevated blood pressure readings, or new use of an anti-hypertensive drug between 18 weeks' gestation and 6 weeks post-partum. The study outcomes were incident CVD and hypertension. Marginal structural Cox models (MSM) were used to account for time-varying confounders and intermediates. Time-fixed exposure defined at the first pregnancy was used in secondary analyses.

RESULTS

A total of 997 women were diagnosed with incident CVD, and 6812 women were diagnosed with hypertension or received a new anti-hypertensive medication during the follow-up period. Compared with women without HDP, those with HDP had a substantially higher rate of CVD (hazard ratio (HR) 2.2, 95% confidence interval (CI) 1.7, 2.7). In women with HDP, the rate of hypertension was five times that of women without a HDP (HR 5.6, 95% CI 5.1, 6.3). With overlapping 95% CIs, the time-fixed analysis and the MSM produced consistent results for both outcomes.

CONCLUSIONS

Women with HDP are at increased risk of developing subsequent CVD and hypertension. Similar estimates obtained with the MSM and the time-fixed analysis suggests that subsequent pregnancies do not confound a first episode of HDP and later CVD.

Social Integration and Reduced Risk of Coronary Heart Disease in Women: The Role of Lifestyle Behaviors

RATIONALE

Higher social integration is associated with lower cardiovascular mortality; however, whether it is associated with incident coronary heart disease (CHD), especially in women, and whether associations differ by case fatality are unclear.

OBJECTIVES

This study sought to examine the associations between social integration and risk of incident CHD in a large female prospective cohort.

METHODS AND RESULTS

Seventy-six thousand three hundred and sixty-two women in the Nurses' Health Study, free of CHD and stroke at baseline (1992), were followed until 2014. Social integration was assessed by a simplified Berkman-Syme Social Network Index every 4 years. End points included nonfatal myocardial infarction and fatal CHD. Two thousand three hundred and seventy-two incident CHD events occurred throughout follow-up. Adjusting for demographic, health/medical risk factors, and depressive symptoms, being socially integrated was significantly associated with lower CHD risk, particularly fatal CHD. The most socially integrated women had a hazard ratio of 0.55 (95% confidence interval, 0.41-0.73) of developing fatal CHD compared with those least socially integrated (P for trend <0.0001). When additionally adjusting for lifestyle behaviors, findings for fatal CHD were maintained but attenuated (P for trend =0.02), whereas the significant associations no longer remained for nonfatal myocardial infarction. The inverse associations between social integration and nonfatal myocardial infarction risk were largely explained by health-promoting behaviors, particularly through differences in cigarette smoking; however, the association with fatal CHD risk remained after accounting for these behaviors and, thus, may involve more direct biological mechanisms.

CONCLUSIONS

Social integration is inversely associated with CHD incidence in women, but is largely explained by lifestyle/behavioral pathways.

A Framework for Treatment Decision Making at Prostate Cancer Recurrence

BACKGROUND

Of the 50,000 men in the US who elect for radical prostatectomy for prostate cancer, 24% to 40% will have a prostate-specific antigen (PSA) recurrence (PSA-R) within 10 years. Deciding whether to administer salvage therapy (ST) at PSA-R presents challenges, as treatment reduces the risk of progression to clinical metastasis but incurs unnecessary side effects should the man die before metastasis. We have developed a new harm-benefit framework using a clinical cohort to inform shared decision making between patients and physicians at PSA-R.

METHODS

Records of 1,045 Johns Hopkins University Hospital patients diagnosed between 1984 and 2013 who had PSA-R following radical prostatectomy were analyzed using marginal structural models to estimate the baseline risk of metastasis and the effect of ST (radiation therapy with or without hormone therapy) while accounting for selection into ST on the basis of PSA growth. The estimated model predicts the harm-benefit tradeoffs of ST within patient subgroups. The benefit of ST is the absolute reduction in the risk of metastasis within 10 years; the harm is the frequency of cancers that would not have metastasized in the patient's lifetime in the absence of ST (overtreatment).

RESULTS

The adjusted hazard ratio associated with ST was 0.41 (95% CI, 0.31 to 0.55). Providing ST to all men at PSA-R reduced the risk of metastasis from 43% to 23% but led to 31% of men being overtreated (harm/benefit = 31/(43-23) = 1.6). Providing ST to men with Gleason score >7 reduced the risk of metastasis from 67% to 39%, with 13% of men being overtreated (harm/benefit = 13/(67-39) = 0.5).

CONCLUSIONS

A quantitative framework that evaluates primary harms and benefits of ST after PSA-R will facilitate informed decision making. Immediate ST may be more appropriate in patient subgroups at elevated risk of metastasis.

Cumulative Effects of Growing Up in Separate and Unequal Neighborhoods on Racial Disparities in Self-rated Health in Early Adulthood

Evidence suggests that living in a socioeconomically deprived neighborhood is associated with worse health. Yet most research relies on cross-sectional data, which implicitly ignore variation in longer-term exposure that may be more consequential for health. Using data from the 1970 to 2011 waves of the Panel Study of Income Dynamics merged with census data on respondents' neighborhoods (N = 1,757), this study estimates a marginal structural model with inverse probability of treatment and censoring weights to examine: (1) whether cumulative exposure to neighborhood disadvantage from birth through age 17 affects self-rated health in early adulthood, and (2) the extent to which variation in such exposure helps to explain racial disparities therein. Findings reveal that prolonged exposure to neighborhood disadvantage throughout childhood and adolescence is strikingly more common among nonwhite versus white respondents and is associated with significantly greater odds of experiencing an incidence of fair or poor health in early adulthood.

Application of the Marginal Structural Model to Account for Suboptimal Adherence in a Randomized Controlled Trial

Background

There is considerable interest in adjusting for suboptimal adherence in randomized controlled trials. A per-protocol analysis, for example removes individuals who fail to achieve a minimal level of adherence. One can also reassign non-adherers to the control group, censor them at the point of non-adherence, or cross them over to the control. However, there are biases inherent in each of these methods. Here, we describe an application of causal modeling to address this issue.

Methods

The marginal structural model with inverse-probability weighting was implemented using a weighted generalized estimating equation model. Two ancillary models were developed to derive the weights. First, stepwise linear regression was used to model the observed percent weight loss, while stepwise logistic regression model was applied to model early discontinuation from the intervention. From these, participant- and time-specific weights were calculated.

Discussion

This model is complicated and requires careful attention to detail. Which variables to force into the ancillary models, how to construct interaction terms, and how to address time-dependent covariates must be considered. Nevertheless, it can be used to great effect to predict intervention effects at full adherence. Moreover, by contrasting these results against intention-to-treat results, insights can be gained into the intrinsic physiologic effect of the intervention.

Trial registration

ClinicalTrials.gov Identifier NCT00427193.

Using Marginal Structural Modeling for Grade Retention Effects

Vandecandelaere, Vansteelandt, De Fraine, and Van Damme (this issue) described marginal structural modeling (MSM) and used it to estimate the effects of a time-varying intervention, retention (holding back) in school grades, on students' math achievement. This commentary supplements Vandecandelaere et al. (this issue) and discusses several topics in retention studies and MSM. First, we discuss the importance of equating time-varying confounders in retention studies. Second, we discuss same-grade and same-age comparisons in retention studies. Third, we discuss one important section in the authors' overview of MSM: why standard methods (e.g., ANCOVA, propensity score analysis) cannot properly adjust for time-varying confounders. Finally, using the grade retention analyses in Vandecandelaere et al. (this issue) as an example, we provide our insights on four aspects of MSM: (a) covariate selection, (b) estimation of weights,

A marginal structural model for recurrent events in the presence of time-dependent confounding: non-specific effects of vaccines on child hospitalisations

Using a Danish register-based study on childhood vaccination and hospitalisation as motivation, a marginal structural model for recurrent events is studied. The model addresses a number of challenges which may be seen more generally in large register-based cohort studies. One is to adjust for a time-dependent confounder when studying the effect of a time-varying exposure on a recurrent event based on an analysis in continuous time. Another is to report results via a measure which is easy to interpret and communicate even though quite elaborate treatment regimes are considered. Lastly, the implementation of continuously updated weights implies a substantial computationally demanding workload. Copyright © 2016 John Wiley & Sons, Ltd.

Statin adherence and risk of acute cardiovascular events among women: a cohort study accounting for time-dependent confounding affected by previous adherence

OBJECTIVES

Previous studies on the effect of statin adherence on cardiovascular events in the primary prevention of cardiovascular disease have adjusted for time-dependent confounding, but potentially introduced bias into their estimates as adherence and confounders were measured simultaneously. We aimed to evaluate the effect when accounting for time-dependent confounding affected by previous adherence as well as time sequence between factors.

DESIGN

Retrospective cohort study.

SETTING

Finnish healthcare registers.

PARTICIPANTS

Women aged 45-64 years initiating statin use for primary prevention of cardiovascular disease in 2001-2004 (n=42 807).

OUTCOMES

Acute cardiovascular event defined as a composite of acute coronary syndrome and acute ischaemic stroke was our primary outcome. Low-energy fractures were used as a negative control outcome to evaluate the healthy-adherer effect.

RESULTS

During the 3-year follow-up, 474 women experienced the primary outcome event and 557 suffered a low-energy fracture. The causal HR estimated with marginal structural model for acute cardiovascular events for all the women who remained adherent (proportion of days covered ≥80%) to statin therapy during the previous adherence assessment year was 0.78 (95% CI: 0.65 to 0.94) when compared with everybody remaining non-adherent (proportion of days covered <80%). The result was robust against alternative model specifications. Statin adherers had a potentially reduced risk of experiencing low-energy fractures compared with non-adherers (HR 0.90, 95% CI 0.76 to 1.07).

CONCLUSIONS

Our study, which took into account the time dependence of adherence and confounders, as well as temporal order between these factors, is support for the concept that adherence to statins in women in primary prevention decreases the risk of acute cardiovascular events by about one-fifth in comparison to non-adherence. However, part of the observed effect of statin adherence on acute cardiovascular events may be due to the healthy-adherer effect.

Longitudinal Study of the Effects of Bacteremia and Sepsis on 5-year Risk of Cardiovascular Events

BACKGROUND

The long-term and cumulative effect of multiple episodes of bacteremia and sepsis across multiple hospitalizations on the development of cardiovascular (CV) events is uncertain.

METHODS

We conducted a longitudinal study of 156 380 hospitalizations in 47 009 patients (≥18 years old) who had at least 2 inpatient admissions at an academic tertiary care center in St Louis, Missouri, from 1 January 2008 through 31 December 2012. We used marginal structural models, estimated by inverse probability weighting (IPW) of bacteremia or sepsis and IPW of censoring, to estimate the marginal causal effects of bacteremia and sepsis on developing the first observed incident CV event, including stroke, transient ischemic attack, and myocardial infarction (MI), during the study period.

RESULTS

Bacteremia and sepsis occurred during 4923 (3.1%) and 5544 (3.5%) hospitalizations among 3932 (8.4%) and 4474 (9.5%) patients, respectively. CV events occurred in 414 (10.5%) and 538 (12.0%) patients with prior episodes of bacteremia or sepsis, respectively, vs 3087 (7.2%) and 2963 (7.0%) patients without prior episodes of bacteremia or sepsis. The causal odds of experiencing a CV event was 1.52-fold (95% confidence interval [CI], 1.21- to 1.90-fold) and 2.39-fold (95% CI, 1.88- to 3.03-fold) higher in patients with prior instances of bacteremia or sepsis, respectively, compared to those without. Prior instances of septic shock resulted in a 6.91-fold (95% CI, 5.34- to 8.93-fold) increase in the odds of MI.

CONCLUSIONS

Prior instances of bacteremia and sepsis substantially increase the 5-year risk of CV events.