Uncertainty directed factorial clinical trials.

The development and evaluation of novel treatment combinations is a key component of modern clinical research. The primary goals of factorial clinical trials of treatment combinations range from the estimation of intervention-specific effects, or the discovery of potential synergies, to the identification of combinations with the highest response probabilities. Most factorial studies use balanced or block randomization, with an equal number of patients assigned to each treatment combination, irrespective of the specific goals of the trial. Here, we introduce a class of Bayesian response-adaptive designs for factorial clinical trials with binary outcomes. The study design was developed using Bayesian decision-theoretic arguments and adapts the randomization probabilities to treatment combinations during the enrollment period based on the available data. Our approach enables the investigator to specify a utility function representative of the aims of the trial, and the Bayesian response-adaptive randomization algorithm aims to maximize this utility function. We considered several utility functions and factorial designs tailored to them. Then, we conducted a comparative simulation study to illustrate relevant differences of key operating characteristics across the resulting designs. We also investigated the asymptotic behavior of the proposed adaptive designs. We also used data summaries from three recent factorial trials in perioperative care, smoking cessation, and infectious disease prevention to define realistic simulation scenarios and illustrate advantages of the introduced trial designs compared to other study designs.

Blood Pressure after Successful Endovascular Therapy: A Systematic Review and Meta-Analysis of Randomized Control Trials.

OBJECTIVE: There are limited data evaluating the optimum blood pressure (BP) goal post mechanical thrombectomy (MT) and its effect on outcomes of patients with large vessel occlusions (LVO). The objective of this study was to compare the efficacy and safety of intensive versus conventional BP control after reperfusion with MT via a systematic review and meta-analysis of randomized controlled trials (RCTs). METHODS: We searched PubMed and Embase to obtain articles related to BP control post MT through September 2023. The primary outcome was functional independence (modified Rankin Scale [mRS] 0-2) at 3 months, while secondary outcomes included excellent outcome (mRS 0-1), symptomatic intracranial hemorrhage (sICH), and mortality. RESULTS: Four RCTs with 1,566 patients (762 randomized into intensive BP control vs. 806 randomized into conventional BP control) were included. Analysis showed that there was a lower likelihood of functional independence (mRS 0-2: odds ratio [OR]: 0.68, 95% confidence interval [CI] 0.51-0.91, p = 0.009) in the more intensive treatment group compared with the conventional treatment group. There was no statistically significant difference in achieving excellent outcome (mRS0-1: OR: 0.82, 95% CI: 0.63-1.07; p = 0.15), risk of sICH or mortality. INTERPRETATION: This systematic review and meta- analysis Indicates that in patients who achieved successful MT for acute ischemic stroke with LVO, intensive BP control was associated with a lower likelihood of functional independence at 3 months without significant difference in likelihood of achieving excellent outcome, sICH risk, or mortality. ANN NEUROL 2024;95:858-865.

Adaptive Randomization Method to Prevent Extreme Instances of Group Size and Covariate Imbalance in Stroke Trials.

BACKGROUND: A recent review of randomization methods used in large multicenter clinical trials within the National Institutes of Health Stroke Trials Network identified preservation of treatment allocation randomness, achievement of the desired group size balance between treatment groups, achievement of baseline covariate balance, and ease of implementation in practice as critical properties required for optimal randomization designs. Common-scale minimal sufficient balance (CS-MSB) adaptive randomization effectively controls for covariate imbalance between treatment groups while preserving allocation randomness but does not balance group sizes. This study extends the CS-MSB adaptive randomization method to achieve both group size and covariate balance while preserving allocation randomness in hyperacute stroke trials. METHODS: A full factorial in silico simulation study evaluated the performance of the proposed new CSSize-MSB adaptive randomization method in achieving group size balance, covariate balance, and allocation randomness compared with the original CS-MSB method. Data from 4 existing hyperacute stroke trials were used to investigate the performance of CSSize-MSB for a range of sample sizes and covariate numbers and types. A discrete-event simulation model created with AnyLogic was used to dynamically visualize the decision logic of the CSSize-MSB randomization process for communication with clinicians. RESULTS: The proposed new CSSize-MSB algorithm uniformly outperformed the CS-MSB algorithm in controlling for group size imbalance while maintaining comparable levels of covariate balance and allocation randomness in hyperacute stroke trials. This improvement was consistent across a distribution of simulated trials with varying levels of imbalance but was increasingly pronounced for trials with extreme cases of imbalance. The results were consistent across a range of trial data sets of different sizes and covariate numbers and types. CONCLUSIONS: The proposed adaptive CSSize-MSB algorithm successfully controls for group size imbalance in hyperacute stroke trials under various settings, and its logic can be readily explained to clinicians using dynamic visualization.

Navigating the landscape of liver cancer management: Study designs in clinical trials and clinical practice.

Hepatocellular carcinoma (HCC) is the fourth leading cause of cancer death worldwide and its prognosis is highly heterogeneous, being related not only to tumour burden but also to the severity of underlying chronic liver disease. Moreover, advances in systemic therapies for HCC have increased the complexity of patient management. Randomised-controlled trials represent the gold standard for evidence generation across all areas of medicine and especially in the oncology field, as they allow for unbiased estimates of treatment effect without confounders. Observational studies have many problems that could reduce their internal and external validity. However, large prospective (well-conducted) observational real-world studies can detect rare adverse events or monitor the occurrence of long-term adverse events. How best to harness real world data, which refers to data generated from the routine care of patients, and real-world 'evidence', which is the evidence generated from real-world data, represents an open challenge. In this review article, we aim to provide an overview of the benefits and limitations of different study designs, particularly focusing on randomised-controlled trials and observational studies, to address important and not fully resolved questions in HCC research.

Lifting the veil off treatment effect heterogeneity.

Clinicians often suspect that a treatment effect can vary across individuals. However, they usually lack "evidence-based" guidance regarding potential heterogeneity of treatment effects (HTE). Potentially actionable HTE is rarely discovered in clinical trials and is widely believed (or rationalized) by researchers to be rare. Conventional statistical methods to test for possible HTE are extremely conservative and tend to reinforce this belief. In truth, though, there is no realistic way to know whether a common, or average, effect estimated from a clinical trial is relevant for all, or even most, patients. This absence of evidence, misinterpreted as evidence of absence, may be resulting in sub-optimal treatment for many individuals. We first summarize the historical context in which current statistical methods for randomized controlled trials (RCTs) were developed, focusing on the conceptual and technical limitations that shaped, and restricted, these methods. In particular, we explain how the common-effect assumption came to be virtually unchallenged. Second, we propose a simple graphical method for exploratory data analysis that can provide useful visual evidence of possible HTE. The basic approach is to display the complete distribution of outcome data rather than relying uncritically on simple summary statistics. Modern graphical methods, unavailable when statistical methods were initially formulated a century ago, now render fine-grained interrogation of the data feasible. We propose comparing observed treatment-group data to "pseudo data" engineered to mimic that which would be expected under a particular HTE model, such as the common-effect model. A clear discrepancy between the distributions of the common-effect pseudo data and the actual treatment-effect data provides prima facie evidence of HTE to motivate additional confirmatory investigation. Artificial data are used to illustrate implications of ignoring heterogeneity in practice and how the graphical method can be useful.

Toward Precision in Critical Care Research: Methods for Observational and Interventional Studies.

Critical care trials evaluate the effect of interventions in patients with diverse personal histories and causes of illness, often under the umbrella of heterogeneous clinical syndromes, such as sepsis or acute respiratory distress syndrome. Given this variation, it is reasonable to expect that the effect of treatment on outcomes may differ for individuals with variable characteristics. However, in randomized controlled trials, efficacy is typically assessed by the average treatment effect (ATE), which quantifies the average effect of the intervention on the outcome in the study population. Importantly, the ATE may hide variations of the treatment's effect on a clinical outcome across levels of patient characteristics, which may erroneously lead to the conclusion that an intervention does not work overall when it may in fact benefit certain patients. In this review, we describe methodological approaches for assessing heterogeneity of treatment effect (HTE), including expert-derived subgrouping, data-driven subgrouping, baseline risk modeling, treatment effect modeling, and individual treatment rule estimation. Next, we outline how insights from HTE analyses can be incorporated into the design of clinical trials. Finally, we propose a research agenda for advancing the field and bringing HTE approaches to the bedside.

The Effectiveness of Lifestyle Interventions in Heart Failure With Preserved Ejection Fraction: A Systematic Review and Network Meta-Analysis.

OBJECTIVES: To perform a network meta-analysis to determine the effectiveness of lifestyle interventions in exercise tolerance and quality of life (QoL) in people with HFpEF. METHODS: Ten databases were searched for randomized controlled trials that evaluated a diet and/or exercise intervention in people with heart failure with preserved ejection fraction until May 2022. The co-primary outcomes were peak oxygen uptake (V̇O2peak) and Quality of Life as assessed by the Minnesota Living with Heart Failure Questionnaire (MLHFQ). We synthesized data using network meta-analysis. RESULTS: We identified 13 trials, including a total of 869 participants, and we incorporated 6 different interventions. Improvements in V̇O2peak compared to controls were seen for all exercise interventions (2.88 [95% CI: 1.36; 4.39] mL/kg/min) for high-intensity interval training (HIIT); 2.37 [95% CI: 1.02; 3.71] mL/kg/min for low-intensity exercise (LIT) combined with a hypocaloric diet; 2.05 [95% CI: 0.81; 3.29] mL/kg/min for moderate-intensity continuous training (MICT); 1.94 [95% CI: 0.59; 3.29] mL/kg/min for LIT; 1.85 [95% CI: 0.27; 3.44] mL/kg/min for MICT combined with resistance training) but not a hypocaloric diet alone (1.26 [95%CI: -0.08; 2.61] mL/kg/min). Only HIIT (-14.45 [95%CI: -24.81; -4.10] points) and LIT (95% CI: -11.05 [-20.55; -1.54] mL/kg/min) significantly improved MLHFQ scores. Network meta-analysis indicated that HIIT was the most effective intervention for improving both V̇O2peak (mean improvement 2.88 [95% CI: 1.36; 4.39] mL/kg/min, follow-up range, 4 weeks-3 years) and QoL (-14.45 [95% CI: -24.81; -4.10] points, follow-up range, 12-26 weeks) compared to usual care. CONCLUSIONS: This network meta-analysis indicates that HIIT is the most effective lifestyle intervention studied to improve exercise capacity and QoL, with mean improvements exceeding the minimum clinically meaningful thresholds. HIIT is likely to be an underused management strategy in HFpEF, but further studies are needed to confirm long-term improvements in symptoms and clinical outcomes.

Rates of bronchopulmonary dysplasia in very low birth weight neonates: a systematic review and meta-analysis.

IMPORTANCE: Large-scale estimates of bronchopulmonary dysplasia (BPD) are warranted for adequate prevention and treatment. However, systematic approaches to ascertain rates of BPD are lacking. OBJECTIVE: To conduct a systematic review and meta-analysis to assess the prevalence of BPD in very low birth weight (≤ 1,500 g) or very low gestational age (< 32 weeks) neonates. DATA SOURCES: A search of MEDLINE from January 1990 until September 2019 using search terms related to BPD and prevalence was performed. STUDY SELECTION: Randomized controlled trials and observational studies evaluating rates of BPD in very low birth weight or very low gestational age infants were eligible. Included studies defined BPD as positive pressure ventilation or oxygen requirement at 28 days (BPD28) or at 36 weeks postmenstrual age (BPD36). DATA EXTRACTION AND SYNTHESIS: Two reviewers independently conducted all stages of the review. Random-effects meta-analysis was used to calculate the pooled prevalence. Subgroup analyses included gestational age group, birth weight group, setting, study period, continent, and gross domestic product. Sensitivity analyses were performed to reduce study heterogeneity. MAIN OUTCOMES AND MEASURES: Prevalence of BPD defined as BPD28, BPD36, and by subgroups. RESULTS: A total of 105 articles or databases and 780,936 patients were included in this review. The pooled prevalence was 35% (95% CI, 28-42%) for BPD28 (n = 26 datasets, 132,247 neonates), and 21% (95% CI, 19-24%) for BPD36 (n = 70 studies, 672,769 neonates). In subgroup meta-analyses, birth weight category, gestational age category, and continent were strong drivers of the pooled prevalence of BPD. CONCLUSIONS AND RELEVANCE: This study provides a global estimation of BPD prevalence in very low birth weight/low gestation neonates.

Effectiveness of pulmonary rehabilitation programmes and/or respiratory muscle training in patients with post-COVID conditions: a systematic review.

BACKGROUND: The term "post-COVID-19 condition" refers to the symptomatology that appears between four to twelve weeks after Covid-19 infection. These symptoms can persist for weeks or even months, significantly diminishing the quality of life for affected individuals. The primary objective of this study was to assess the effectiveness of pulmonary rehabilitation programs and/or respiratory muscle training on respiratory sequelae in patients with post-COVID condition. METHODS: The literature search was conducted in the following databases: PubMed, PEDro, Embase, Cochrane, Scopus, and Web of Science. Randomized clinical trials were included in which participants were aged 18 years or older. Articles were excluded if at least one of the therapies did not involve pulmonary rehabilitation or respiratory muscle training, if the participants were COVID positive, if studies lacked results, and finally, if interventions were conducted without supervision or at home. This review only encompasses supervised non-virtual interventions. This study adheres to the PRISMA statement and has been registered in the PROSPERO database (CRD42023433843). RESULTS: The outcomes obtained in the included studies are assessed across the following variables: Exercise capacity using the 6-minute walk test, Dyspnea, fatigue, Pulmonary function, Maximum inspiratory pressure, and Quality of life. CONCLUSION: Despite the absence of a specific treatment at present, it was evident from this review that a well-structured pulmonary rehabilitation program that incorporates both aerobic and muscular strength exercises along with techniques and inspiratory muscle exercises was the most effective form of treatment.

Frailty as an Effect Modifier in Randomized Controlled Trials: A Systematic Review.

BACKGROUND: The effect of clinical interventions may vary by patients' frailty status. Understanding treatment effect heterogeneity by frailty could lead to frailty-guided treatment strategies and reduce overtreatment and undertreatment. This systematic review aimed to examine the effect modification by frailty in randomized controlled trials (RCTs) that evaluate pharmacological, non-pharmacological, and multicomponent interventions. METHODS: We searched PubMed, Web of Science, EMBASE, and ClinicalTrial.gov, from their inception to 8 December 2023. Two reviewers independently extracted trial data and examined the study quality with senior authors. RESULTS: Sixty-one RCTs that evaluated the interaction between frailty and treatment effects in older adults were included. Frailty was evaluated using different tools such as the deficit accumulation frailty index, frailty phenotype, and other methods. The effect of several pharmacological interventions (e.g., edoxaban, sacubitril/valsartan, prasugrel, and chemotherapy) varied according to the degree of frailty, whereas other treatments (e.g., antihypertensives, vaccinations, osteoporosis medications, and androgen medications) demonstrated consistent benefits across different frailty levels. Some non-pharmacological interventions had greater benefits in patients with higher (e.g., chair yoga, functional walking, physical rehabilitation, and higher dose exercise program) or lower (e.g., intensive lifestyle intervention, psychosocial intervention) levels of frailty, while others (e.g., resistance-type exercise training, moderate-intensive physical activity, walking and nutrition or walking) produced similar intervention effects. Specific combined interventions (e.g., hospital-based disease management programs) demonstrated inconsistent effects across different frailty levels. DISCUSSION: The efficacy of clinical interventions often varied by frailty levels, suggesting that frailty is an important factor to consider in recommending clinical interventions in older adults. REGISTRATION: PROSPERO registration number CRD42021283051.

Effectiveness of Very Brief Advice on Tobacco Cessation: A Systematic Review and Meta-Analysis.

BACKGROUND: Very brief advice (VBA; ≤ 3 min) on quitting is practical and scalable during brief medical interactions with patients who smoke. This study aims to synthesize the effectiveness of VBA for smoking cessation and summarize the implementation strategies. METHODS: We searched randomized controlled trials aiming at tobacco abstinence and comparing VBA versus no smoking advice or no contact from Medline, Embase, CINAHL, Cochrane Library, PsycInfo databases, six Chinese databases, two trial registries ClinicalTrials.gov and WHO-ICTRP from inception to September 30, 2023. Grading of Recommendations, Assessment, Development, and Evaluations framework was used to assess the certainty of the evidence of the meta-analytic findings. The outcomes were self-reported long-term tobacco abstinence at least 6 months after treatment initiation, earlier than 6 months after treatment initiation, and quit attempts. Effect sizes were computed as risk ratio (RR) with 95% CI using frequentist random-effect models. DATA SYNTHESIS: Thirteen randomized controlled trials from 15 articles (n = 26,437) were included. There was moderate-certainty evidence that VBA significantly increased self-reported tobacco abstinence at ≥ 6 months in the adjusted model (adjusted risk ratio ARR 1.17, 95% CI: 1.07-1.27) compared with controls. The sensitivity analysis showed similar results when abstinence was verified by biochemical validation (n = 6 studies, RR 1.53, 95% CI 0.98-2.40). There was high-certainty evidence that VBA significantly increased abstinence at < 6 months (ARR 1.22, 95% CI: 1.01-1.47). Evidence of effect on quit attempts (ARR 1.03, 95% CI 0.97-1.08) was of very low certainty. DISCUSSION: VBA delivered in a clinical setting is effective in increasing self-reported tobacco abstinence, which provides support for wider adoption in clinical practice.

Identifying the Best Initial Oral Antibiotics for Adults with Community-Acquired Pneumonia: A Network Meta-Analysis.

BACKGROUND: The objective of this network meta-analysis was to compare rates of clinical response and mortality for empiric oral antibiotic regimens in adults with mild-moderate community-acquired pneumonia (CAP). METHODS: We searched PubMed, Cochrane, and the reference lists of systematic reviews and clinical guidelines. We included randomized trials of adults with radiologically confirmed mild to moderate CAP initially treated orally and reporting clinical cure or mortality. Abstracts and studies were reviewed in parallel for inclusion in the analysis and for data abstraction. We performed separate analyses by antibiotic medications and antibiotic classes and present the results through network diagrams and forest plots sorted by p-scores. We assessed the quality of each study using the Cochrane Risk of Bias framework, as well as global and local inconsistency. RESULTS: We identified 24 studies with 9361 patients: six at low risk of bias, six at unclear risk, and 12 at high risk. Nemonoxacin, levofloxacin, and telithromycin were most likely to achieve clinical response (p-score 0.79, 0.71, and 0.69 respectively), while penicillin and amoxicillin were least likely to achieve clinical response. Levofloxacin, nemonoxacin, azithromycin, and amoxicillin-clavulanate were most likely to be associated with lower mortality (p-score 0.85, 0.75, 0.74, and 0.68 respectively). By antibiotic class, quinolones and macrolides were most effective for clinical response (0.71 and 0.70 respectively), with amoxicillin-clavulanate plus macrolides and beta-lactams being less effective (p-score 0.11 and 0.22). Quinolones were most likely to be associated with lower mortality (0.63). All confidence intervals were broad and partially overlapping. CONCLUSION: We observed trends toward a better clinical response and lower mortality for quinolones as empiric antibiotics for CAP, but found no conclusive evidence of any antibiotic being clearly more effective than another. More trials are needed to inform guideline recommendations on the most effective antibiotic regimens for outpatients with mild to moderate CAP.

The methodological quality of 176,620 randomized controlled trials published between 1966 and 2018 reveals a positive trend but also an urgent need for improvement.

Many randomized controlled trials (RCTs) are biased and difficult to reproduce due to methodological flaws and poor reporting. There is increasing attention for responsible research practices and implementation of reporting guidelines, but whether these efforts have improved the methodological quality of RCTs (e.g., lower risk of bias) is unknown. We, therefore, mapped risk-of-bias trends over time in RCT publications in relation to journal and author characteristics. Meta-information of 176,620 RCTs published between 1966 and 2018 was extracted. The risk-of-bias probability (random sequence generation, allocation concealment, blinding of patients/personnel, and blinding of outcome assessment) was assessed using a risk-of-bias machine learning tool. This tool was simultaneously validated using 63,327 human risk-of-bias assessments obtained from 17,394 RCTs evaluated in the Cochrane Database of Systematic Reviews (CDSR). Moreover, RCT registration and CONSORT Statement reporting were assessed using automated searches. Publication characteristics included the number of authors, journal impact factor (JIF), and medical discipline. The annual number of published RCTs substantially increased over 4 decades, accompanied by increases in authors (5.2 to 7.8) and institutions (2.9 to 4.8). The risk of bias remained present in most RCTs but decreased over time for allocation concealment (63% to 51%), random sequence generation (57% to 36%), and blinding of outcome assessment (58% to 52%). Trial registration (37% to 47%) and the use of the CONSORT Statement (1% to 20%) also rapidly increased. In journals with a higher impact factor (>10), the risk of bias was consistently lower with higher levels of RCT registration and the use of the CONSORT Statement. Automated risk-of-bias predictions had accuracies above 70% for allocation concealment (70.7%), random sequence generation (72.1%), and blinding of patients/personnel (79.8%), but not for blinding of outcome assessment (62.7%). In conclusion, the likelihood of bias in RCTs has generally decreased over the last decades. This optimistic trend may be driven by increased knowledge augmented by mandatory trial registration and more stringent reporting guidelines and journal requirements. Nevertheless, relatively high probabilities of bias remain, particularly in journals with lower impact factors. This emphasizes that further improvement of RCT registration, conduct, and reporting is still urgently needed.

Application of the Estimand Framework to Anesthesia Trials.

SUMMARY: Events occurring after randomization, such as use of rescue medication, treatment discontinuation, or death, are common in randomized trials. These events can change either the existence or interpretation of the outcome of interest. However, appropriate handling of these intercurrent events is often unclear. The International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) E9(R1) addendum introduced the estimand framework, which aligns trial objectives with the design, conduct, statistical analysis, and interpretation of results. This article describes how the estimand framework can be used in anesthesia trials to precisely define the treatment effect to be estimated, key attributes of an estimand, common intercurrent events in anesthesia trials with strategies for handling them, and use of the estimand framework in a hypothetical anesthesia trial on postoperative delirium. When planning anesthesia trials, clearly defining the estimand is vital to ensure that what is being estimated is clearly understood, is clinically relevant, and helps answer the clinical questions of interest.

Sequential covariate-adjusted randomization via hierarchically minimizing Mahalanobis distance and marginal imbalance.

In comparative studies, covariate balance and sequential allocation schemes have attracted growing academic interest. Although many theoretically justified adaptive randomization methods achieve the covariate balance, they often allocate patients in pairs or groups. To better meet the practical requirements where the clinicians cannot wait for other participants to assign the current patient for some economic or ethical reasons, we propose a method that randomizes patients individually and sequentially. The proposed method conceptually separates the covariate imbalance, measured by the newly proposed modified Mahalanobis distance, and the marginal imbalance, that is the sample size difference between the 2 groups, and it minimizes them with an explicit priority order. Compared with the existing sequential randomization methods, the proposed method achieves the best possible covariate balance while maintaining the marginal balance directly, offering us more control of the randomization process. We demonstrate the superior performance of the proposed method through a wide range of simulation studies and real data analysis, and also establish theoretical guarantees for the proposed method in terms of both the convergence of the imbalance measure and the subsequent treatment effect estimation.

On exact randomization-based covariate-adjusted confidence intervals.

Randomization-based inference using the Fisher randomization test allows for the computation of Fisher-exact P-values, making it an attractive option for the analysis of small, randomized experiments with non-normal outcomes. Two common test statistics used to perform Fisher randomization tests are the difference-in-means between the treatment and control groups and the covariate-adjusted version of the difference-in-means using analysis of covariance. Modern computing allows for fast computation of the Fisher-exact P-value, but confidence intervals have typically been obtained by inverting the Fisher randomization test over a range of possible effect sizes. The test inversion procedure is computationally expensive, limiting the usage of randomization-based inference in applied work. A recent paper by Zhu and Liu developed a closed form expression for the randomization-based confidence interval using the difference-in-means statistic. We develop an important extension of Zhu and Liu to obtain a closed form expression for the randomization-based covariate-adjusted confidence interval and give practitioners a sufficiency condition that can be checked using observed data and that guarantees that these confidence intervals have correct coverage. Simulations show that our procedure generates randomization-based covariate-adjusted confidence intervals that are robust to non-normality and that can be calculated in nearly the same time as it takes to calculate the Fisher-exact P-value, thus removing the computational barrier to performing randomization-based inference when adjusting for covariates. We also demonstrate our method on a re-analysis of phase I clinical trial data.

A generalized outcome-adaptive sequential multiple assignment randomized trial design.

A dynamic treatment regime (DTR) is a mathematical representation of a multistage decision process. When applied to sequential treatment selection in medical settings, DTRs are useful for identifying optimal therapies for chronic diseases such as AIDs, mental illnesses, substance abuse, and many cancers. Sequential multiple assignment randomized trials (SMARTs) provide a useful framework for constructing DTRs and providing unbiased between-DTR comparisons. A limitation of SMARTs is that they ignore data from past patients that may be useful for reducing the probability of exposing new patients to inferior treatments. In practice, this may result in decreased treatment adherence or dropouts. To address this problem, we propose a generalized outcome-adaptive (GO) SMART design that adaptively unbalances stage-specific randomization probabilities in favor of treatments observed to be more effective in previous patients. To correct for bias induced by outcome adaptive randomization, we propose G-estimators and inverse-probability-weighted estimators of DTR effects embedded in a GO-SMART and show analytically that they are consistent. We report simulation results showing that, compared to a SMART, Response-Adaptive SMART and SMART with adaptive randomization, a GO-SMART design treats significantly more patients with the optimal DTR and achieves a larger number of total responses while maintaining similar or better statistical power.

Do machine learning methods lead to similar individualized treatment rules? A comparison study on real data.

Identifying patients who benefit from a treatment is a key aspect of personalized medicine, which allows the development of individualized treatment rules (ITRs). Many machine learning methods have been proposed to create such rules. However, to what extent the methods lead to similar ITRs, that is, recommending the same treatment for the same individuals is unclear. In this work, we compared 22 of the most common approaches in two randomized control trials. Two classes of methods can be distinguished. The first class of methods relies on predicting individualized treatment effects from which an ITR is derived by recommending the treatment evaluated to the individuals with a predicted benefit. In the second class, methods directly estimate the ITR without estimating individualized treatment effects. For each trial, the performance of ITRs was assessed by various metrics, and the pairwise agreement between all ITRs was also calculated. Results showed that the ITRs obtained via the different methods generally had considerable disagreements regarding the patients to be treated. A better concordance was found among akin methods. Overall, when evaluating the performance of ITRs in a validation sample, all methods produced ITRs with limited performance, suggesting a high potential for optimism. For non-parametric methods, this optimism was likely due to overfitting. The different methods do not lead to similar ITRs and are therefore not interchangeable. The choice of the method strongly influences for which patients a certain treatment is recommended, drawing some concerns about their practical use.

Adaptive designs in public health: Vaccine and cluster randomized trials go Bayesian.

Clinical trials in public health-particularly those conducted in low- and middle-income countries-often involve communicable and non-communicable diseases with high disease burden and unmet needs. Trials conducted in these regions often are faced with resource limitations, so improving the efficiencies of these trials is critical. Adaptive trial designs have the potential to save trial time and resources and reduce the number of patients receiving ineffective interventions. In this paper, we provide a detailed account of the implementation of vaccine and cluster randomized trials within the framework of Bayesian adaptive trials, with emphasis on computational efficiency and flexibility with regard to stopping rules and allocation ratios. We offer an educated approach to selecting prior distributions and a data-driven empirical Bayes method for plug-in estimates for nuisance parameters.