Search

In both observational studies and randomized trials with noncompliance, unmeasured confounding may exist which may bias treatment effect estimates. Instrumental variables (IV) are a popular technique for addressing such confounding, enabling consistent estimation of causal effects. This paper proposes nonparametric IV estimators for censored time to event data that may be subject to competing risks. A simple, plug-in estimator is introduced using nonparametric estimators of the cumulative incidence function, with confidence intervals derived using asymptotic theory. To provide an overall test of the treatment effect, an integrated weighted difference statistic is suggested, which is applicable to data with and without competing risks. Simulation studies demonstrate that the methods perform well with realistic samples sizes. The methods are applied to assess the effect of infant or maternal antiretroviral therapy on transmission of HIV from mother to child via breastfeeding using data from a large, recently completed randomized trial in Malawi where noncompliance with assigned treatment may confound treatment effect estimates. [ABSTRACT FROM AUTHOR]

Background: We examined associations between suicidality and genotypes that predict plasma efavirenz exposure among AIDS Clinical Trials Group study participants in the United States.Methods: Four clinical trials randomly assigned treatment-naive participants to efavirenz-containing regimens; suicidality was defined as reported suicidal ideation or attempted or completed suicide. Genotypes that predict plasma efavirenz exposure were defined by CYP2B6 and CYP2A6 polymorphisms. Associations were evaluated with weighted Cox proportional hazards models stratified by race/ethnicity. Additional analyses adjusted for genetic ancestry and selected covariates.Results: Among 1833 participants, suicidality was documented in 41 in exposed analyses, and 34 in on-treatment analyses. In unadjusted analyses based on 12 genotype levels, suicidality increased per level in exposed (hazard ratio, 1.11; 95% confidence interval, .96-1.27) and on-treatment 1.16; 1.01-1.34) analyses. In the on-treatment analysis, the association was strongest among white but nearly null among black participants. Considering 3 metabolizer levels (extensive, intermediate and slow), slow metabolizers were at increased risk. Results were similar after baseline covariate-adjustment for genetic ancestry, sex, age, weight, injection drug use history, and psychiatric history or recent psychoactive medication.Conclusions: Genotypes that predict higher plasma efavirenz exposure were associated with increased risk of suicidality. Strength of association varied by race/ethnicity. [ABSTRACT FROM AUTHOR]