Faculty Bibliography

Missing outcome data is a crucial threat to the validity of treatment effect estimates from randomized trials. The outcome distributions of participants with missing and observed data are often different, which increases bias. Causal inference methods may aid in reducing the bias and improving efficiency by incorporating baseline variables into the analysis. In particular, doubly robust estimators incorporate 2 nuisance parameters: the outcome regression and the missingness mechanism (ie, the probability of missingness conditional on treatment assignment and baseline variables), to adjust for differences in the observed and unobserved groups that can be explained by observed covariates. To consistently estimate the treatment effect, one of these nuisance parameters must be consistently estimated. Traditionally, nuisance parameters are estimated using parametric models, which often precludes consistency, particularly in moderate to high dimensions. Recent research on missing data has focused on data-adaptive estimation to help achieve consistency, but the large sample properties of such methods are poorly understood. In this article, we discuss a doubly robust estimator that is consistent and asymptotically normal under data-adaptive estimation of the nuisance parameters. We provide a formula for an asymptotically exact confidence interval under minimal assumptions. We show that our proposed estimator has smaller finite-sample bias compared to standard doubly robust estimators. We present a simulation study demonstrating the enhanced performance of our estimators in terms of bias, efficiency, and coverage of the confidence intervals. We present the results of an illustrative example: a randomized, double-blind phase 2/3 trial of antiretroviral therapy in HIV-infected persons.

BACKGROUND:Patients at intermediate risk (IR) according to The Society of Thoracic Surgeons risk score are today frequently oriented toward the transfemoral aortic valve replacement (TAVR) option. Our goal was to evaluate the best treatment strategies for IR patients with severe aortic stenosis. METHODS:Of a consecutive series of 1,144 surgical aortic valve replacements (AVRs) performed in our institution between 2008 and 2014, we reviewed the early and late outcomes of two different groups: a low-risk (LR) group of 470 patients, and an IR group of 620. We eliminated from the analysis 54 high-risk patients who were currently candidates for TAVR. All patients underwent surgical AVR with or without concomitant coronary artery bypass grafting. Social Security database interrogation provided long-term information. RESULTS:The early mortality rate (30 days) between LR and IR patients was similar (1.70% vs 2.74%, p = 0.25) and both lower than predicted mortality rates. However, cumulative 5-year survival was significantly higher in LR patients (86.3%) than in IR patients (75.4%; p = 0.0007 by log-rank test), although excellent in IR group. Comparing IR survivors and nonsurvivors, ages at operation were 69.5 ± 12.7 years for survivors vs 75.4 ± 9.6 years for those experiencing late deaths (p = 0.002). Risk factors for late deaths after multivariate analysis were age, hemodialysis, and chronic lung disease. CONCLUSIONS:Most IR patients today should undergo surgical AVR, but because of survival rates combined with still unavailable late structural deterioration rates in TAVR valves, patients in the IR group with high Society of Thoracic Surgeons scores and known risk factors may be better served with TAVR as data regarding late percutaneous valve function accrue.