Interim monitoring in sequential multiple assignment randomized trials

A sequential multiple assignment randomized trial (SMART) facilitates the comparison of multiple adaptive treatment strategies (ATSs) simultaneously. Previous studies have established a framework to test the homogeneity of multiple ATSs by a global Wald test through inverse probability weighting. SMARTs are generally lengthier than classical clinical trials due to the sequential nature of treatment randomization in multiple stages. Thus, it would be beneficial to add interim analyses allowing for an early stop if overwhelming efficacy is observed. We introduce group sequential methods to SMARTs to facilitate interim monitoring based on the multivariate chi-square distribution. Simulation studies demonstrate that the proposed interim monitoring in SMART (IM-SMART) maintains the desired type I error and power with reduced expected sample size compared to the classical SMART. Finally, we illustrate our method by reanalyzing a SMART assessing the effects of cognitive behavioral and physical therapies in patients with knee osteoarthritis and comorbid subsyndromal depressive symptoms.     

Optimal adaptive randomized designs for clinical trials

Optimal decision-analytic designs are deterministic. Such designsare appropriately criticized in the context of clinical trialsbecause they are subject to assignment bias. On the other hand,balanced randomized designs may assign an excessive number ofpatients to a treatment arm that is performing relatively poorly.We propose a compromise between these two extremes, one thatachieves some of the good characteristics of both. We introducea constrained optimal adaptive design for a fully sequentialrandomized clinical trial with k arms and n patients. An r-designis one for which, at each allocation, each arm has probabilityat least r of being chosen, 0 r 1/k. An optimal design amongall r-designs is called r-optimal. An r₁-design is also an r₂-designif r₁ r₂. A design without constraint is the special case r = 0and a balanced randomized design is the special case r = 1/k.The optimization criterion is to maximize the expected overallutility in a Bayesian decision-analytic approach, where utilityis the sum over the utilities for individual patients over a‘patient horizon’ N. We prove analytically thatthere exists an r-optimal design such that each patient is assignedto a particular one of the arms with probability 1 – (k – 1)r,and to the remaining arms with probability r. We also show thatthe balanced design is asymptotically r-optimal for any givenr, 0 r < 1/k, as N/n  . This implies that everyr-optimal design is asymptotically optimal without constraint.Numerical computations using backward induction for k = 2arms show that, in general, this asymptotic optimality featurefor r-optimal designs can be accomplished with moderate trialsize n if the patient horizon N is large relative to n. We alsoshow that, in a trial with an r-optimal design, r < 1/2,fewer patients are assigned to an inferior arm than when followinga balanced design, even for r-optimal designs having the samestatistical power as a balanced design. We discuss extensionsto various clinical trial settings.     

Stochastic inequality probabilities for adaptively randomized clinical trials

We examine stochastic inequality probabilities of the form P (X > Y) and P (X > max (Y, Z)) where X, Y, and Z are random variables with beta, gamma, or inverse gamma distributions. We discuss the applications of such inequality probabilities to adaptively randomized clinical trials as well as methods for calculating their values.     

Community-equipoise and the ethics of randomized clinical trials

This paper critically examines a particular strategy for resolving the central ethical dilemma associated with randomized clincial trials (RCTs) -- the "community equipoise" strategy (CE). The dilemma is that RCTs appear to violate a physician's duty to choose that therapy which there is most reason to believe is in the patient's best interest, randomizing patients even once evidence begins to favor one treatment. The community equipose strategy involves the suggestion that our judgment that neither treatment is to be preferred (that there obtains a state of "equipoise") is to be assessed according to a community rather than an individual standard. Thus, though a physician may personally believe that there is some reason to prefer one treatment, patients can legitimately be randomized if there remains disagreement in the community of medical professionals. Rationales in favor of this conception include the following: (i) medical knowledge is best understood as residing in the community, (ii) the judgments of others count as evidence, and so should change one's own opinion, (iii) subjects would not be better off outside the trial, and (iv) the point of any trial is the resolution of dispute in the medical community. I critically examine these rationales and argue that they are insufficient. Amongst the problems are tensions between various of these underlying rationales, and important ambiguities in just what the CE criterion is to amount to. Finally, I argue that even if use of CE was justified, it would not justify carrying out RCTs anywhere near long enough to discharge our duty to gain reliable knowledge on which to base safe and effective medical practice. Hence, we need some different justification for carrying out RCTs.     

Multiple imputation methods for treatment noncompliance and nonresponse in randomized clinical trials

Summary .  Randomized clinical trials are a powerful tool for investigating causal treatment effects, but in human trials there are oftentimes problems of noncompliance which standard analyses, such as the intention-to-treat or as-treated analysis, either ignore or incorporate in such a way that the resulting estimand is no longer a causal effect. One alternative to these analyses is the complier average causal effect (CACE) which estimates the average causal treatment effect among a subpopulation that would comply under any treatment assigned. We focus on the setting of a randomized clinical trial with crossover treatment noncompliance (e.g., control subjects could receive the intervention and intervention subjects could receive the control) and outcome nonresponse. In this article, we develop estimators for the CACE using multiple imputation methods, which have been successfully applied to a wide variety of missing data problems, but have not yet been applied to the potential outcomes setting of causal inference. Using simulated data we investigate the finite sample properties of these estimators as well as of competing procedures in a simple setting. Finally we illustrate our methods using a real randomized encouragement design study on the effectiveness of the influenza vaccine.     

A mixed model-based variance estimator for marginal model analyses of cluster randomized trials

Generalized estimating equations (GEE) are used in the analysis of cluster randomized trials (CRTs) because: 1) the resulting intervention effect estimate has the desired marginal or population-averaged interpretation, and 2) most statistical packages contain programs for GEE. However, GEE tends to underestimate the standard error of the intervention effect estimate in CRTs. In contrast, penalized quasi-likelihood (PQL) estimates the standard error of the intervention effect in CRTs much better than GEE but is used less frequently because: 1) it generates an intervention effect estimate with a conditional, or cluster-specific, interpretation, and 2) PQL is not a part of most statistical packages. We propose taking the variance estimator from PQL and re-expressing it as a sandwich-type estimator that could be easily incorporated into existing GEE packages, thereby making GEE useful for the analysis of CRTs. Using numerical examples and data from an actual CRT, we compare the performance of this variance estimator to others proposed in the literature, and we find that our variance estimator performs as well as or better than its competitors.     

Adjustment for exploratory cut-off selection in randomized clinical trials with survival endpoint

Defining the target population based on predictive biomarkers plays an important role during clinical development. After establishing a relationship between a biomarker candidate and response to treatment in exploratory phases, a subsequent confirmatory trial ideally involves only subjects with high potential of benefiting from the new compound. In order to identify those subjects in case of a continuous biomarker, a cut-off is needed. Usually, a cut-off is chosen that resulted in a subgroup with a large observed treatment effect in an exploratory trial. However, such a data-driven selection may lead to overoptimistic expectations for the subsequent confirmatory trial. Treatment effect estimates, probability of success, and posterior probabilities are useful measures for deciding whether or not to conduct a confirmatory trial enrolling the biomarker-defined population. These measures need to be adjusted for selection bias. We extend previously introduced Approximate Bayesian Computation techniques for adjustment of subgroup selection bias to a time-to-event setting with cut-off selection. Challenges in this setting are that treatment effects become time-dependent and that subsets are defined by the biomarker distribution. Simulation studies show that the proposed method provides adjusted statistical measures which are superior to naïve Maximum Likelihood estimators as well as simple shrinkage estimators.     

Adaptive design and estimation in randomized clinical trials with correlated observations

Clinical trial designs involving correlated data often arise in biomedical research. The intracluster correlation needs to be taken into account to ensure the validity of sample size and power calculations. In contrast to the fixed-sample designs, we propose a flexible trial design with adaptive monitoring and inference procedures. The total sample size is not predetermined, but adaptively re-estimated using observed data via a systematic mechanism. The final inference is based on a weighted average of the block-wise test statistics using generalized estimating equations, where the weight for each block depends on cumulated data from the ongoing trial. When there are no significant treatment effects, the devised stopping rule allows for early termination of the trial and acceptance of the null hypothesis. The proposed design updates information regarding both the effect size and within-cluster correlation based on the cumulated data in order to achieve a desired power. Estimation of the parameter of interest and its confidence interval are proposed. We conduct simulation studies to examine the operating characteristics and illustrate the proposed method with an example.     

Representing supraspecific taxa in higher-level phylogenetic analyses: guidelines for palaeontologists

Abstract:  As phylogenetic analyses become larger, one of the greatest methodological difficulties is representing speciose supraspecific clades in higher-level analyses (e.g. trilobites within studies of arthropod phylogeny). Several strategies have been proposed, including using representative single composite terminals or species-level exemplars, and various methods are currently used in the palaeontological literature. However, this is problematic, as simulation studies and empirical arguments in the systematics literature have clearly identified multiple exemplars as the optimal method. The continuing usage of suboptimal strategies in palaeontology may lessen the accuracy of phylogenies and hampers comparison between alternative studies. Here, I outline problems with suboptimal strategies, review arguments in support of multiple exemplars and provide guidelines for palaeontologists undertaking higher-level phylogenetic analyses.     

Designs for clinical trials with time-to-event outcomes based on stopping guidelines for lack of benefit

Background

We present the design, methods and population characteristics of a large community trial that assessed the efficacy of a weekly supplement containing vitamin A or beta-carotene, at recommended dietary levels, in reducing maternal mortality from early gestation through 12 weeks postpartum. We identify challenges faced and report solutions in implementing an intervention trial under low-resource, rural conditions, including the importance of population choice in promoting generalizability, maintaining rigorous data quality control to reduce inter- and intra- worker variation, and optimizing efficiencies in information and resources flow from and to the field.

Methods

This trial was a double-masked, cluster-randomized, dual intervention, placebo-controlled trial in a contiguous rural area of ~435 sq km with a population of ~650,000 in Gaibandha and Rangpur Districts of Northwestern Bangladesh. Approximately 120,000 married women of reproductive age underwent 5-weekly home surveillance, of whom ~60,000 were detected as pregnant, enrolled into the trial and gave birth to ~44,000 live-born infants. Upon enrollment, at ~ 9 weeks' gestation, pregnant women received a weekly oral supplement containing vitamin A (7000 ug retinol equivalents (RE)), beta-carotene (42 mg, or ~7000 ug RE) or a placebo through 12 weeks postpartum, according to prior randomized allocation of their cluster of residence. Systems described include enlistment and 5-weekly home surveillance for pregnancy based on menstrual history and urine testing, weekly supervised supplementation, periodic risk factor interviews, maternal and infant vital outcome monitoring, birth defect surveillance and clinical/biochemical substudies.

Results

The primary outcome was pregnancy-related mortality assessed for 3 months following parturition. Secondary outcomes included fetal loss due to miscarriage or stillbirth, infant mortality under three months of age, maternal obstetric and infectious morbidity, infant infectious morbidity, maternal and infant micronutrient status, fetal and infant growth and prematurity, external birth defects and postnatal infant growth to 3 months of age.

Conclusion

Aspects of study site selection and its "resonance" with national and rural qualities of Bangladesh, the trial's design, methods and allocation group comparability achieved by randomization, field procedures and innovative approaches to solving challenges in trial conduct are described and discussed. This trial is registered with http://Clinicaltrials.gov as protocol NCT00198822.     

Acupuncture for functional recovery after stroke: a systematic review of sham-controlled randomized clinical trials

Background

Acupuncture is frequently advocated as an adjunct treatment during stroke rehabilitation. The aim of this review was to assess its effectiveness in this setting.

Methods

We searched 25 databases and 12 major Korean traditional medicine journals from their inception to October 2009. We included randomized controlled trials, with no language restrictions, that compared the effects of acupuncture (with or without electrical stimulation) with sham acupuncture. We assessed the methodologic quality of the trials using the Cochrane risk-of-bias criteria and the PEDro (Physiotherapy Evidence Database) scale.

Results

Ten of 664 potentially relevant studies met our inclusion criteria. For acute and subacute stages after stroke, we included seven trials. A meta-analysis of the five studies that assessed functionality did not show a significant difference in favour of acupuncture, with high heterogeneity. A post-hoc sensitivity analysis of three trials with low risk of bias did not show beneficial effects of acupuncture on activities of daily living at the end of the intervention period (n = 244; standard mean difference 0.07, 95% confidence interval [CI] −0.18 to 0.32; I² = 0%) or after follow-up (n = 244; standard mean difference 0.10, 95% CI −0.15 to 0.35; I² = 0%). For the chronic stage after stroke, three trials tested effects of acupuncture on function according to the Modified Ashworth Scale; all failed to show favourable effects.

Interpretation

Our meta-analyses of data from rigorous randomized sham-controlled trials did not show a positive effect of acupuncture as a treatment for functional recovery after stroke.Acupuncture is often used as an adjunct to mainstream rehabilitation after stroke. It involves the insertion of an acupuncture needle into the skin at certain points of the body. Acupuncture is claimed to be effective for a wide range of conditions, such as pain, musculoskeletal disorders and several neurologic diseases.¹ Possible mechanisms of its effects on neurologic conditions include stimulation of neuronal cell proliferation,² facilitation of neural plasticity,³ reduction of the post-ischemic inflammatory reaction⁴ and prevention of neuronal apoptosis.⁵Before acupuncture can be recommended for routine use, we require evidence from rigorous randomized clinical trials. In acupuncture trials, it is difficult to allow blinding of the treatment allocation.⁶ A placebo must be indistinguishable from the real treatment and inert.⁶ “Sham” is used to describe any control procedure that is used to blind treatment allocation in clinical trials of acupuncture.⁶ Several sham procedures are now available, such as the use of penetrating acupuncture on nonacupuncture points, superficial puncture of the skin on acupuncture points and nonpenetration on acupuncture points with sham needle devices.⁶Several reviews assessing the effects of acupuncture for stroke have been published. However, some did not include all of the relevant articles published in Asian countries; others included interventions other than acupuncture; and several were not systematic. We conducted a systematic review to critically evaluate all of the currently available randomized sham-controlled trials of acupuncture as an adjunct to mainstream stroke rehabilitation.     

Interim discharge summaries: how are they best delivered to general practitioners?

All patients discharged from a medical ward during four months were randomly assigned to one of two groups. In one group the patients were given their interim discharge summary for delivery to their general practitioner by hand; in the other group the summary was posted by the hospital. Of the 289 summaries sent by either method, 279 (97%) arrived at the general practitioner''s surgery. A mean (median) time of two (one) days elapsed before arrival when summaries were delivered by hand and a mean (median) of four and a half (four) days when they were posted; at least 55% of summaries delivered by hand arrived within one day of the day of discharge compared with 8% of those posted. If all interim discharge summaries were given to patients to deliver communication between hospitals and general practitioners would be accelerated and considerable savings might be made.