Power analysis in a SMART design: sample size estimation for determining the best embedded dynamic treatment regime

Abstract

Sequential, multiple assignment, randomized trial (SMART) designs have become increasingly popular in the field of precision medicine by providing a means for comparing sequences of treatments tailored to the individual patient, i.e., dynamic treatment regime (DTR). The construction of evidence-based DTRs promises a replacement to adhoc one-size-fits-all decisions pervasive in patient care. However, there are substantial statistical challenges in sizing SMART designs due to the complex correlation structure between the DTRs embedded in the design. Since the primary goal of SMARTs is the construction of an optimal DTR, investigators are interested in sizing SMARTs based on the ability to screen out DTRs inferior to the optimal DTR by a given amount which cannot be done using existing methods. In this paper, we fill this gap by developing a rigorous power analysis framework that leverages multiple comparisons with the best methodology. Our method employs Monte Carlo simulation in order to compute the minimum number of individuals to enroll in an arbitrary SMART. We will evaluate our method through extensive simulation studies. We will illustrate our method by retrospectively computing the power in the Extending Treatment Effectiveness of Naltrexone SMART study.

SMARTp: A SMART design for nonsurgical treatments of chronic periodontitis with spatially referenced and nonrandomly missing skewed outcomes

“…To the best of our knowledge, this is the first SMART proposal in CP research within the umbrella of precision oral health —a major goal in the NIH/NIDCR's Strategic Plan 2014–2019, and advances previous SMART proposals (Ghosh et al., ; NeCamp et al., ) considered for clustered data. Very recently, Artman, Nahum‐Shani, Wu, Mckay, and Ertefaie () proposed rigorous sample size and power calculations factoring in the complex correlation structure between the DTRs that remain embedded within the SMART by design. However, their setup is different than what we are considering here.…”

“…() and Artman et al. (), the proposed sample size method can also be extended to detect the optimal treatment regime. These are important avenues for future research, and will be considered elsewhere.…”

“…Furthermore, the efficiency of the proposed estimator can be improved using the corresponding doubly robust estimators, for example, those considered by Ertefaie et al (2015). Following Ertefaie et al (2015) and Artman et al (2018), the proposed sample size method can also be extended to detect the optimal treatment regime. These are important avenues for future research, and will be considered elsewhere.…”

How to determine whether conceptual endophenotypes can improve clinical outcomes in patients suffering from major depression: An exploratory approach

“…Methods and concepts underpinning treatment selection research have made progress (e.g., Cohen and DeRubeis, 2018). Methods to plan such studies with appropriate sample sizes, both with view to confirmatory trials (sample size calculation/power analysis: Artman et al, 2018) as well as studies in naturalistic settings or secondary data analyses (Riley et al, in press a, b) are increasingly available. And due to the high number of potential confounding variables including social-psychological aspects such as expectations on clinicians' and patients' side, and non-specific treatment factors (Lambert, 2013), planning such studies is a particular challenge.…”

Personalized Dual-Hormone Control for Type 1 Diabetes Using Deep Reinforcement Learning

“…However, several reasons hinder the building of efficient RL models to solve problems in chronic diseases. Firstly, as RL medical data is collected from a dynamic interaction between the human and environment, they are limited and expensive (Artman et al 2018). In addition, it is different to a scenario such as playing Atari Copyright c 2020, Association for the Advancement of Artificial Intelligence (www.aaai.org).…”