A framework for testing non-inferiority in a three-arm, sequential, multiple assignment randomized trial

Sequential multiple assignment randomized trial design is becoming increasingly used in the field of precision medicine. This design allows comparisons of sequences of adaptive interventions tailored to the individual patient. Superiority testing is usually the initial goal in order to determine which embedded adaptive intervention yields the best primary outcome on average. When direct superiority is not evident, yet an adaptive intervention poses other benefits, then non-inferiority testing is warranted. Non-inferiority testing in the sequential multiple assignment randomized trial setup is rather new and involves the specification of non-inferiority margin and other important assumptions that are often unverifiable internally. These challenges are not specific to sequential multiple assignment randomized trial and apply to two-arm non-inferiority trials that do not include a standard-of-care (or placebo) arm. To address some of these challenges, three-arm non-inferiority trials that include the standard-of-care arm are proposed. However, methods developed so far for three-arm non-inferiority trials are not sequential multiple assignment randomized trial-specific. This is because apart from embedded adaptive interventions, sequential multiple assignment randomized trial typically does not include a third standard-of-care arm. In this article, we consider a three-arm sequential multiple assignment randomized trial from an National Institutes of Health-funded study of symptom management strategies among people undergoing cancer treatment. Motivated by that example, we propose a novel data analytic method for non-inferiority testing in the framework of three-arm sequential multiple assignment randomized trial for the first time. Sample size and power considerations are discussed through extensive simulation studies to elucidate our method.

Modified interactive Q-learning for attenuating the impact of model misspecification with treatment effect heterogeneity

A sequential multiple assignment randomized trial, which incorporates multiple stages of randomization, is a popular approach for collecting data to inform personalized and adaptive treatments. There is an extensive literature on statistical methods to analyze data collected in sequential multiple assignment randomized trials and estimate the optimal dynamic treatment regime. Q-learning with linear regression is widely used for this purpose due to its ease of implementation. However, model misspecification is a common problem with this approach, and little attention has been given to the impact of model misspecification when treatment effects are heterogeneous across subjects. This article describes the integrative impact of two possible types of model misspecification related to treatment effect heterogeneity: omitted early-stage treatment effects in late-stage main effect model, and violated linearity assumption between pseudo-outcomes and predictors despite non-linearity arising from the optimization operation. The proposed method, aiming to deal with both types of misspecification concomitantly, builds interactive models into modified parametric Q-learning with Murphy's regret function. Simulations show that the proposed method is robust to both sources of model misspecification. The proposed method is applied to a two-stage sequential multiple assignment randomized trial with embedded tailoring aimed at reducing binge drinking in first-year college students.

Bayesian sample size calculations for comparing two strategies in SMART studies

In the management of most chronic conditions characterized by the lack of universally effective treatments, adaptive treatment strategies (ATSs) have grown in popularity as they offer a more individualized approach. As a result, sequential multiple assignment randomized trials (SMARTs) have gained attention as the most suitable clinical trial design to formalize the study of these strategies. While the number of SMARTs has increased in recent years, sample size and design considerations have generally been carried out in frequentist settings. However, standard frequentist formulae require assumptions on interim response rates and variance components. Misspecifying these can lead to incorrect sample size calculations and correspondingly inadequate levels of power. The Bayesian framework offers a straightforward path to alleviate some of these concerns. In this paper, we provide calculations in a Bayesian setting to allow more realistic and robust estimates that account for uncertainty in inputs through the 'two priors' approach. Additionally, compared to the standard frequentist formulae, this methodology allows us to rely on fewer assumptions, integrate pre-trial knowledge, and switch the focus from the standardized effect size to the MDD. The proposed methodology is evaluated in a thorough simulation study and is implemented to estimate the sample size for a full-scale SMART of an internet-based adaptive stress management intervention on cardiovascular disease patients using data from its pilot study conducted in two Canadian provinces.

Efficient and robust approaches for analysis of sequential multiple assignment randomized trials: Illustration using the ADAPT-R trial

Personalized intervention strategies, in particular those that modify treatment based on a participant's own response, are a core component of precision medicine approaches. Sequential multiple assignment randomized trials (SMARTs) are growing in popularity and are specifically designed to facilitate the evaluation of sequential adaptive strategies, in particular those embedded within the SMART. Advances in efficient estimation approaches that are able to incorporate machine learning while retaining valid inference can allow for more precise estimates of the effectiveness of these embedded regimes. However, to the best of our knowledge, such approaches have not yet been applied as the primary analysis in SMART trials. In this paper, we present a robust and efficient approach using targeted maximum likelihood estimation (TMLE) for estimating and contrasting expected outcomes under the dynamic regimes embedded in a SMART, together with generating simultaneous confidence intervals for the resulting estimates. We contrast this method with two alternatives (G-computation and inverse probability weighting estimators). The precision gains and robust inference achievable through the use of TMLE to evaluate the effects of embedded regimes are illustrated using both outcome-blind simulations and a real-data analysis from the Adaptive Strategies for Preventing and Treating Lapses of Retention in Human Immunodeficiency Virus (HIV) Care (ADAPT-R) trial (NCT02338739), a SMART with a primary aim of identifying strategies to improve retention in HIV care among people living with HIV in sub-Saharan Africa.

Outcome trajectory estimation for optimal dynamic treatment regimes with repeated measures

In recent sequential multiple assignment randomized trials, outcomes were assessed multiple times to evaluate longer-term impacts of the dynamic treatment regimes (DTRs). Q-learning requires a scalar response to identify the optimal DTR. Inverse probability weighting may be used to estimate the optimal outcome trajectory, but it is inefficient, susceptible to model mis-specification, and unable to characterize how treatment effects manifest over time. We propose modified Q-learning with generalized estimating equations to address these limitations and apply it to the M-bridge trial, which evaluates adaptive interventions to prevent problematic drinking among college freshmen. Simulation studies demonstrate our proposed method improves efficiency and robustness.

Sample size estimation for comparing dynamic treatment regimens in a SMART: A Monte Carlo-based approach and case study with longitudinal overdispersed count outcomes

Dynamic treatment regimens (DTRs), also known as treatment algorithms or adaptive interventions, play an increasingly important role in many health domains. DTRs are motivated to address the unique and changing needs of individuals by delivering the type of treatment needed, when needed, while minimizing unnecessary treatment. Practically, a DTR is a sequence of decision rules that specify, for each of several points in time, how available information about the individual's status and progress should be used in practice to decide which treatment (e.g. type or intensity) to deliver. The sequential multiple assignment randomized trial (SMART) is an experimental design widely used to empirically inform the development of DTRs. Sample size planning resources for SMARTs have been developed for continuous, binary, and survival outcomes. However, an important gap exists in sample size estimation methodology for SMARTs with longitudinal count outcomes. Furthermore, in many health domains, count data are overdispersed-having variance greater than their mean. We propose a Monte Carlo-based approach to sample size estimation applicable to many types of longitudinal outcomes and provide a case study with longitudinal overdispersed count outcomes. A SMART for engaging alcohol and cocaine-dependent patients in treatment is used as motivation.

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.

A Sequential Multiple Assignment Randomized Trial of Symptom Management After Chemotherapy

CONTEXT

Many cancer survivors experience a lingering symptom burden after chemotherapy.

OBJECTIVES

In this sequential multiple assignment randomized trial, we tested optimal sequencing of two evidence-based interventions for symptom management.

METHODS

Survivors of solid tumors (N = 451) were interviewed at baseline and stratified as high or low need for symptom management based on comorbidity and depressive symptoms. High need survivors were randomized initially to the 12-week Symptom Management and Survivorship Handbook (SMSH, N = 282) or 12-week SMSH with eight weeks of Telephone Interpersonal Counseling (TIPC, N = 93) added during weeks one to eight. After four weeks of the SMSH alone, non-responders on depression were re-randomized to continue with SMSH alone (N = 30) or add TIPC (N = 31). Severity of depression and summed severity index of 17 other symptoms over weeks one to13 were compared between randomized groups and among three dynamic treatment regimes (DTRs): 1) SMSH for 12 weeks; 2) SMSH for 12 weeks with eight weeks of TIPC from week one; 3) SMSH for four weeks followed by SMSH+TIPC for eight weeks if no response to the SMSH alone on depression at week four.

RESULTS

There were no main effects for randomized arms or DTRs, but there was a significant interaction of trial arm with baseline depression favoring SMSH alone during weeks one to four in the first randomization and SMSH+TIPC in the second randomization.

CONCLUSION

The SMSH may represent a simple effective option for symptom management, adding TIPC only when there is no response to SMSH alone for people with elevated depression and multiple co-morbidities.

Adaptive intervention for prevention of adolescent suicidal behavior after hospitalization: a pilot sequential multiple assignment randomized trial

BACKGROUND

The need for effective interventions for psychiatrically hospitalized adolescents who have varying levels of postdischarge suicide risk calls for personalized approaches, such as adaptive interventions (AIs). We conducted a nonrestricted pilot Sequential, Multiple Assignment, Randomized Trial (SMART) to guide the development of an AI targeting suicide risk after hospitalization.

METHODS

Adolescent inpatients (N = 80; ages 13-17; 67.5% female) were randomized in Phase 1 to a Motivational Interview-Enhanced Safety Plan (MI-SP), delivered during hospitalization, alone or in combination with postdischarge text-based support (Texts). Two weeks after discharge, participants were re-randomized in Phase 2 to added telephone booster calls or to no calls. Mechanisms of change were assessed with daily diaries for four weeks and over a 1- and 3-month follow-up. This trial is registered with clinicaltrials.gov (identifier: NCT03838198).

RESULTS

Procedures were feasible and acceptable. Mixed effects models indicate that adolescents randomized to MI-SP + Texts (Phase 1) and those randomized to booster calls (Phase 2) experienced significant improvement in daily-level mechanisms, including safety plan use, self-efficacy to refrain from suicidal action, and coping by support seeking. Those randomized to MI-SP + Texts also reported significantly higher coping self-efficacy at 1 and 3 months. Although exploratory, results were in the expected direction for MI-SP + Texts, versus MI-SP alone, in terms of lower risk of suicide attempts (Hazard ratio = 0.30; 95% CI = 0.06, 1.48) and suicidal behavior (Hazard ratio = 0.36; 95% CI = 0.10, 1.37) three months after discharge. Moreover, augmentation with booster calls did not have an overall meaningful impact on suicide attempts (Hazard ratio = 0.65; 95% CI = 0.17, 3.05) or suicidal behavior (Hazard ratio = 0.78; 95% CI = 0.23, 2.67); however, boosters benefited most those initially assigned to MI-SP + Texts.

CONCLUSIONS

The current SMART was feasible and acceptable for the purpose of informing an AI for suicidal adolescents, warranting additional study. Findings also indicate that postdischarge text-based support offers a promising augmentation to safety planning delivered during hospitalization.

Study Designs to Assess Real-World Interventions to Prevent COVID-19

Background: In the face of the novel virus SARS-CoV-2, scientists and the public are eager for evidence about what measures are effective at slowing its spread and preventing morbidity and mortality. Other than mathematical modeling, studies thus far evaluating public health and behavioral interventions at scale have largely been observational and ecologic, focusing on aggregate summaries. Conclusions from these studies are susceptible to bias from threats to validity such as unmeasured confounding, concurrent policy changes, and trends over time. We offer recommendations on how to strengthen frequently applied study designs which have been used to understand the impact of interventions to reduce the spread of COVID-19, and suggest implementation-focused, pragmatic designs that, moving forward, could be used to build a robust evidence base for public health practice. Methods: We conducted a literature search of studies that evaluated the effectiveness of non-pharmaceutical interventions and policies to reduce spread, morbidity, and mortality of COVID-19. Our targeted review of the literature aimed to explore strengths and weaknesses of implemented studies, provide recommendations for improvement, and explore alternative real-world study design methods to enhance evidence-based decision-making. Results:Study designs such as pre/post, interrupted time series, and difference-in-differences have been used to evaluate policy effects at the state or country level of a range of interventions, such as shelter-in-place, face mask mandates, and school closures. Key challenges with these designs include the difficulty of disentangling the effects of contemporaneous changes in policy and correctly modeling infectious disease dynamics. Pragmatic study designs such as the SMART (Sequential, Multiple-Assignment Randomized Trial), stepped wedge, and preference designs could be used to evaluate community re-openings such as schools, and other policy changes. Conclusions: As the epidemic progresses, we need to move from post-hoc analyses of available data (appropriate for the beginning of the pandemic) to proactive evaluation to ensure the most rigorous approaches possible to evaluate the impact of COVID-19 prevention interventions. Pragmatic study designs, while requiring initial planning and community buy-in, could offer more robust evidence on what is effective and for whom to combat the global pandemic we face and future policy decisions.

Dynamic treatment regimes, past, present, and future: A conversation with experts

We asked three leading researchers in the area of dynamic treatment regimes to share their stories on how they became interested in this topic and their perspectives on the most important opportunities and challenges for the future.

Comparison of adaptive treatment strategies based on longitudinal outcomes in sequential multiple assignment randomized trials

In sequential multiple assignment randomized trials, longitudinal outcomes may be the most important outcomes of interest because this type of trials is usually conducted in areas of chronic diseases or conditions. We propose to use a weighted generalized estimating equation (GEE) approach to analyzing data from such type of trials for comparing two adaptive treatment strategies based on generalized linear models. Although the randomization probabilities are known, we consider estimated weights in which the randomization probabilities are replaced by their empirical estimates and prove that the resulting weighted GEE estimator is more efficient than the estimators with true weights. The variance of the weighted GEE estimator is estimated by an empirical sandwich estimator. The time variable in the model can be linear, piecewise linear, or more complicated forms. This provides more flexibility that is important because, in the adaptive treatment setting, the treatment changes over time and, hence, a single linear trend over the whole period of study may not be practical. Simulation results show that the weighted GEE estimators of regression coefficients are consistent regardless of the specification of the correlation structure of the longitudinal outcomes. The weighted GEE method is then applied in analyzing data from the Clinical Antipsychotic Trials of Intervention Effectiveness. Copyright © 2016 John Wiley & Sons, Ltd.

Adaptive Interventions and SMART Designs: Application to child behavior research in a community setting

Heterogeneity between and within people necessitates the need for sequential personalized interventions to optimize individual outcomes. Personalized or adaptive interventions (AIs) are relevant for diseases and maladaptive behavioral trajectories when one intervention is not curative and success of a subsequent intervention may depend on individual characteristics or response. AIs may be applied to medical settings and to investigate best prevention, education, and community-based practices. AIs can begin with low-cost or low-burden interventions and followed with intensified or alternative interventions for those who need it most. AIs that guide practice over the course of a disease, program, or school year can be investigated through sequential multiple assignment randomized trials (SMARTs). To promote the use of SMARTs, we provide a hypothetical SMART in a Head Start program to address child behavior problems. We describe the advantages and limitations of SMARTs, particularly as they may be applied to the field of evaluation.

Comparing dynamic treatment regimes using repeated-measures outcomes: modeling considerations in SMART studies

A dynamic treatment regime (DTR) is a sequence of decision rules, each of which recommends a treatment based on a patient's past and current health status. Sequential, multiple assignment, randomized trials (SMARTs) are multi-stage trial designs that yield data specifically for building effective DTRs. Modeling the marginal mean trajectories of a repeated-measures outcome arising from a SMART presents challenges, because traditional longitudinal models used for randomized clinical trials do not take into account the unique design features of SMART. We discuss modeling considerations for various forms of SMART designs, emphasizing the importance of considering the timing of repeated measures in relation to the treatment stages in a SMART. For illustration, we use data from three SMART case studies with increasing level of complexity, in autism, child attention deficit hyperactivity disorder, and adult alcoholism. In all three SMARTs, we illustrate how to accommodate the design features along with the timing of the repeated measures when comparing DTRs based on mean trajectories of the repeated-measures outcome.

A "SMART" design for building individualized treatment sequences

Interventions often involve a sequence of decisions. For example, clinicians frequently adapt the intervention to an individual's outcomes. Altering the intensity and type of intervention over time is crucial for many reasons, such as to obtain improvement if the individual is not responding or to reduce costs and burden when intensive treatment is no longer necessary. Adaptive interventions utilize individual variables (severity, preferences) to adapt the intervention and then dynamically utilize individual outcomes (response to treatment, adherence) to readapt the intervention. The Sequential Multiple Assignment Randomized Trial (SMART) provides high-quality data that can be used to construct adaptive interventions. We review the SMART and highlight its advantages in constructing and revising adaptive interventions as compared to alternative experimental designs. Selected examples of SMART studies are described and compared. A data analysis method is provided and illustrated using data from the Extending Treatment Effectiveness of Naltrexone SMART study.