Search

In some cluster randomized trials, subjects may not comply with their assigned treatment status. Such treatment noncompliance can create opportunities for spillover effects within clusters. Little research has focused on what can be learned in such context. This article provides a non-technical review of recent work on the complications that arise in cluster randomized trials where some units within treated clusters do not comply with treatment but the treatment spillovers over to these units. We motivate concepts using a hypothetical vaccine cluster randomized trial. We review that standard instrumental variable methods cannot recover the complier average causal effect in the presence of these spillovers. In fact, we review that without additional assumptions, little can be learned about compliance effects or spillover effects. We discuss one additional assumption that allows for bounds on a key causal effect. We also outline an estimator for these bounds. [ABSTRACT FROM AUTHOR]

We investigate the efficacy of surgical versus nonsurgical management for two gastrointestinal conditions, colitis and diverticulitis, using observational data. We deploy an instrumental variable design with surgeons' tendencies to operate as an instrument. Assuming instrument validity, we find that nonsurgical alternatives can reduce both hospital length of stay and the risk of complications, with estimated effects larger for septic patients than for nonseptic patients. The validity of our instrument is plausible but not ironclad, necessitating a sensitivity analysis. Existing sensitivity analyses for IV designs assume effect homogeneity, unlikely to hold here because of patient-specific physiology. We develop a new sensitivity analysis that accommodates arbitrary effect heterogeneity and exploits components explainable by observed features. We find that the results for nonseptic patients prove more robust to hidden bias despite having smaller estimated effects. For nonseptic patients, two individuals with identical observed characteristics would have to differ in their odds of assignment to a high tendency to operate surgeon by a factor of 2.34 to overturn our finding of a benefit for nonsurgical management in reducing length of stay. For septic patients, this value is only 1.64. Simulations illustrate that this phenomenon may be explained by differences in within-group heterogeneity. for this article, including a standardized description of the materials available for reproducing the work, are available as an online supplement. [ABSTRACT FROM AUTHOR]

Instrumental variable (IV) analyses are becoming common in health services research and epidemiology. Most IV analyses use naturally occurring instruments, such as distance to a hospital. In these analyses, investigators must assume that the instrument is as-if randomly assigned. This assumption cannot be tested directly, but it can be falsified. Most IV falsification tests compare relative prevalence or bias in observed covariates between the instrument and exposure. These tests require investigators to make covariate-by-covariate judgments about the validity of the IV design. Often, only some covariates are well-balanced, making it unclear whether as-if randomization can be assumed for the instrument. We propose an alternative falsification test that compares IV balance or bias with the balance or bias that would have been produced under randomization. A key advantage of our test is that it allows for global balance measures as well as easily interpretable graphical comparisons. Furthermore, our test does not rely on parametric assumptions and can be used to validly assess whether the instrument is significantly closer to being as-if randomized than the exposure. We demonstrate our approach using data from (SPOT)light, a prospective cohort study carried out in 48 National Health Service hospitals in the United Kingdom between November 1, 2010, and December 31, 2011. This study used bed availability in the intensive care unit as an instrument for admission to the intensive care unit. [ABSTRACT FROM AUTHOR]

Summary: Instrumental variable methods, subject to appropriate identification assumptions, enable consistent estimation of causal effects in the presence of unobserved confounding. Near–far matching has been proposed as one analytic method to improve inference by strengthening the effect of the instrument on the exposure and balancing observable characteristics between groups of subjects with low and high values of the instrument. However, in settings with hierarchical data (e.g. patients nested within hospitals), or where several covariate interactions must be balanced, conventional near–far matching algorithms may fail to achieve the requisite covariate balance. We develop a new matching algorithm, that combines near–far matching with refined covariate balance, to balance large numbers of nominal covariates while also strengthening the instrumental variable. This extension of near–far matching is motivated by a case‐study that aims to identify the causal effect of prompt admission to an intensive care unit on 7‐day and 28‐day mortality. [ABSTRACT FROM AUTHOR]

Pretreatment selection or censoring ("selection on treatment") can occur when two treatment levels are compared ignoring the third option of neither treatment, in "censoring by death" settings where treatment is only defined for those who survive long enough to receive it, or in general in studies where the treatment is only defined for a subset of the population. Unfortunately, the standard instrumental variable (IV) estimand is not defined in the presence of such selection, so we consider estimating a new survivor-complier causal effect. Although this effect is generally not identified under standard IV assumptions, it is possible to construct sharp bounds. We derive these bounds and give a corresponding data-driven sensitivity analysis, along with nonparametric yet efficient estimation methods. Importantly, our approach allows for high-dimensional confounding adjustment, and valid inference even after employing machine learning. Incorporating covariates can tighten bounds dramatically, especially when they are strong predictors of the selection process. We apply the methods in a UK cohort study of critical care patients to examine the mortality effects of prompt admission to the intensive care unit, using ICU bed availability as an instrument. Supplementary materials for this article are available online. [ABSTRACT FROM AUTHOR]

Confounding by indication is a critical challenge in evaluating the effectiveness of surgical interventions using observational data. The threat from confounding is compounded when using medical claims data due to the inability to measure risk severity. If there are unobserved differences in risk severity across patients, treatment effect estimates based on methods such a multivariate regression may be biased in an unknown direction. A research design based on instrumental variables offers one possibility for reducing bias from unobserved confounding compared to risk adjustment with observed confounders. This study investigates whether a physician's preference for operative care is a valid instrumental variable for studying the effect of emergency surgery. We review the plausibility of the necessary causal assumptions in an investigation of the effect of emergency general surgery (EGS) on inpatient mortality among adults using medical claims data from Florida, Pennsylvania, and New York in 2012–2013. In a departure from the extant literature, we use the framework of stochastic monotonicity which is more plausible in the context of a preference-based instrument. We compare estimates from an instrumental variables design to estimates from a design based on matching that assumes all confounders are observed. Estimates from matching show lower mortality rates for patients that undergo EGS compared to estimates based in the instrumental variables framework. Results vary substantially by condition type. We also present sensitivity analyses as well as bounds for the population level average treatment effect. We conclude with a discussion of the interpretation of estimates from both approaches. [ABSTRACT FROM AUTHOR]

Summary: The method of instrumental variables provides a framework to study causal effects in both randomized experiments with non‐compliance and in observational studies where natural circumstances produce as if random nudges to accept treatment. Traditionally, inference for instrumental variables relied on asymptotic approximations of the distribution of the Wald estimator or two‐stage least squares, often with structural modelling assumptions and/or moment conditions. We utilize the randomization inference approach to instrumental variables inference. First, we outline the exact method, which uses the randomized assignment of treatment in experiments as a basis for inference but lacks a closed form solution and may be computationally infeasible in many applications. We then provide an alternative to the exact method, the almost exact method, which is computationally feasible but retains the advantages of the exact method. We also review asymptotic methods of inference, including those associated with two‐stage least squares, and analytically compare them with randomization inference methods. We also perform additional comparisons by using a set of simulations. We conclude with three different applications from the social sciences. [ABSTRACT FROM AUTHOR]

In randomized controlled trials with non-adherence, instrumental variable (IV) methods are frequently used to report the complier average causal effect. With binary outcomes, many of the available IV estimation methods impose distributional assumptions. We develop a randomization-inference-based method of IV estimation for binary outcomes. The method is non-parametric and is based on Fisher's exact test, and estimates can be easily calculated from a set of 2×2 or 2×2×2 tables. Although we retain the standard IV identification assumptions for confidence regions and point estimates, the IV estimand under randomization inference is sample specific and does not assume that the randomized controlled trials participants are a random sample from the target population. We illustrate the method with the 'IMPROVE' trial that compares emergency endovascular versus open surgical repair for patients with ruptured aortic aneurysms. [ABSTRACT FROM AUTHOR]

Causal analysis in program evaluation has primarily focused on the question about whether or not a program, or package of policies, has an impact on the targeted outcome of interest. However, it is often of scientific and practical importance to also explain why such impacts occur. In this paper, we introduce causal mediation analysis, a statistical framework for analyzing causal mechanisms that has become increasingly popular in social and medical sciences in recent years. The framework enables us to show exactly what assumptions are sufficient for identifying causal mediation effects for the mechanisms of interest, derive a general algorithm for estimating such mechanism-specific effects, and formulate a sensitivity analysis for the violation of those identification assumptions. We also discuss an extension of the framework to analyze causal mechanisms in the presence of treatment noncompliance, a common problem in randomized evaluation studies. The methods are illustrated via applications to two intervention studies on pre-school classes and job-training workshops. [ABSTRACT FROM AUTHOR]