An open-label pilot study of quetiapine plus mirtazapine for heavy drinkers with alcohol use disorder.

Animal research suggests that medications that produce a weak dopamine D2 receptor blockade and potentiate noradrenergic activity may decrease alcohol drinking. In an open-label pilot study of subjects with alcohol dependence, we tested whether the combination of quetiapine, a weak dopamine D2 receptor antagonist, whose primary metabolite, desalkylquetiapine, is a norepinephrine reuptake inhibitor, and mirtazapine, a potent α2 norepinephrine receptor antagonist, would decrease alcohol drinking and craving. Twenty very heavy drinkers with alcohol dependence entered a trial of 8 weeks of treatment with quetiapine followed by 8 weeks of treatment with a combination of quetiapine plus mirtazapine. Alcohol use was assessed weekly with a Timeline Follow-Back interview and craving with the Penn Alcohol Craving Scale. Among the 11 completers, subjects reported improved outcomes in the quetiapine plus mirtazapine period compared to the quetiapine alone period: fewer very heavy drinking days per week (1.3 [SD = 2.4] vs. 2.1 [SD = 2.8]; t = 2.3, df = 10, p = 0.04); fewer total number of drinks per week (39.7 [SD = 61.6] vs. 53.4 [SD = 65.0]; t = 2.8, df = 10, p = 0.02); and lower craving scores (2.5 [SD = 1.4] vs. 3.2 [SD = 1.2]; t = 2.4, df = 10, p = 0.04). All subjects reported at least one adverse event; 72.7% reported somnolence. In this open-label pilot study, treatment with quetiapine plus mirtazapine was associated with a decrease in alcohol drinking and craving. These findings are consistent with our previous work in animal models of alcohol use disorders and suggest that further study of medications or combinations of medications with this pharmacologic profile is warranted.

Desipramine enhances the ability of paliperidone to decrease alcohol drinking.

Alcohol use disorder commonly occurs in patients with schizophrenia and dramatically worsens their course. The atypical antipsychotic clozapine has been associated with reduced drinking in these patients, but its toxicity reduces its use. We have attempted to create a clozapine-like drug by combining agents that capture components of clozapine's pharmacologic action, including its weak dopamine D2 blockade and noradrenergic modulation. The current study assessed whether paliperidone, a dopamine D2 receptor and adrenergic alpha-2 receptor antagonist like clozapine, would attenuate alcohol drinking in the alcohol-preferring P rat and the Syrian golden hamster, and whether desipramine, a norepinephrine reuptake inhibitor, would potentiate the ability of paliperidone to attenuate alcohol drinking in the P rat and the Syrian golden hamster. Daily subcutaneous injections of paliperidone (5 mg/kg for the rat; 1 mg/kg for the hamster) over 20 days slightly and transiently attenuated initiation of alcohol consumption in both animals. Desipramine (3 mg/kg) or lower doses of paliperidone alone did not affect alcohol drinking. However, the combination of desipramine (3 mg/kg) and paliperidone essentially prevented initiation of alcohol drinking and acquisition of alcohol preference in the P rat (2.5 or 5 mg/kg), and almost as dramatically suppressed chronic alcohol intake and alcohol preference in the hamster (2.5 mg/kg). Taken together, the current data suggest that (1) the desipramine and paliperidone combination attenuates alcohol drinking in a synergistic manner, and (2) desipramine and paliperidone may serve as an effective new treatment for alcohol use disorder in patients with schizophrenia.

Long-acting injectable vs oral risperidone for schizophrenia and co-occurring alcohol use disorder: a randomized trial.

OBJECTIVE: Alcohol use disorders worsen the course of schizophrenia. Although the atypical antipsychotic clozapine appears to decrease alcohol use in schizophrenia, risperidone does not. We have proposed that risperidone's relatively potent dopamine D2 receptor blockade may partly underlie its lack of effect on alcohol use. Since long-acting injectable (LAI) risperidone both results in lower average steady-state plasma concentrations than oral risperidone (with lower D2 receptor occupancy) and encourages adherence, it may be more likely to decrease heavy alcohol use (days per week of drinking 5 or more drinks per day) than oral risperidone. METHOD: Ninety-five patients with DSM-IV-TR diagnoses of schizophrenia and alcohol use disorder were randomized to 6 months of oral or LAI risperidone between 2005 and 2008. Explanatory (efficacy) analyses were carried out to evaluate the potential benefits of LAI under suitably controlled conditions (in contrast to real-world settings), with intent-to-treat analyses being secondary. RESULTS: Explanatory analyses showed that heavy drinking in the oral group worsened over time (P = .024) and that there was a statistical trend toward significance in the difference between the changes in heavy drinking days in the oral and LAI groups (P = .054). Furthermore, the 2 groups differed in the mean number of drinking days per week (P = .035). The intent-to-treat analyses showed no difference in heavy drinking but did show a difference in average drinking days per week similar to that obtained from the explanatory analyses (P = .018). Neither explanatory nor intent-to-treat analyses showed any between-group differences in alcohol use as measured by intensity or the Alcohol Use Scale. The plasma concentrations of the active metabolite 9-hydroxyrisperidone were significantly lower in patients taking LAI (P < .05), despite their significantly (overall) better treatment adherence (P < .005). CONCLUSION: For the population considered here, schizophrenia patients with alcohol use disorder appear to continue drinking some alcohol while taking either form of risperidone. Nonetheless, our data suggest that injectable risperidone may be a better choice than the oral form for these dual diagnosis patients. TRIAL REGISTRATION: ClinicalTrials.gov identifier: NCT00130923.

Clozapine reconstructed: Haloperidol's ability to reduce alcohol intake in the Syrian golden hamster can be enhanced through noradrenergic modulation by desipramine and idazoxan.

BACKGROUND: Alcohol use disorder commonly occurs in patients with schizophrenia. Most antipsychotic drugs do not lessen alcohol use; although the atypical antipsychotic clozapine has been shown to reduce alcohol use in patients with schizophrenia, its toxicity severely limits its use in patients. With an eye toward creation of a safer clozapine-like drug, we have investigated the pharmacological basis of the clozapine's effects on alcohol drinking in the Syrian golden hamster. In this animal, as in patients with schizophrenia, clozapine reduces alcohol drinking while the typical antipsychotic haloperidol does not. We have suggested that clozapine decreases alcohol drinking due to its weak dopamine D2 receptor blockade, its potent norepinephrine α-2 receptor antagonism, as well as its ability to elevate plasma norepinephrine. METHODS: We recreated a clozapine-like drug to reduce alcohol drinking in the Syrian golden hamster by combining low dose haloperidol with a norepinephrine α-2 receptor antagonist, idazoxan, and a norepinephrine reuptake inhibitor, desipramine. Hamsters were given free access to water and alcohol (15% v/v) and were treated daily with each drug or with the three-drug combination for 23 days. RESULTS: The drug combination reduced alcohol drinking and preference significantly as compared to vehicle or to haloperidol, idazoxan or desipramine, while not altering food-intake or body-weight. CONCLUSION: These findings suggest that that haloperidol, which does not reduce alcohol drinking in patients with schizophrenia or the hamster, if combined with idazoxan and desipramine (producing a drug combination that mimics aspects of clozapine's pharmacology) is able to reduce alcohol drinking in the hamster.

Design and inference for the intent-to-treat principle using adaptive treatment.

Nonadherence to assigned treatment jeopardizes the power and interpretability of intent-to-treat comparisons from clinical trial data and continues to be an issue for effectiveness studies, despite their pragmatic emphasis. We posit that new approaches to design need to complement developments in methods for causal inference to address nonadherence, in both experimental and practice settings. This paper considers the conventional study design for psychiatric research and other medical contexts, in which subjects are randomized to treatments that are fixed throughout the trial and presents an alternative that converts the fixed treatments into an adaptive intervention that reflects best practice. The key element is the introduction of an adaptive decision point midway into the study to address a patient's reluctance to remain on treatment before completing a full-length trial of medication. The clinical uncertainty about the appropriate adaptation prompts a second randomization at the new decision point to evaluate relevant options. Additionally, the standard 'all-or-none' principal stratification (PS) framework is applied to the first stage of the design to address treatment discontinuation that occurs too early for a midtrial adaptation. Drawing upon the adaptive intervention features, we develop assumptions to identify the PS causal estimand and to introduce restrictions on outcome distributions to simplify expectation-maximization calculations. We evaluate the performance of the PS setup, with particular attention to the role played by a binary covariate. The results emphasize the importance of collecting covariate data for use in design and analysis. We consider the generality of our approach beyond the setting of psychiatric research.

Measuring the quality of dying and death in the pediatric intensive care setting: the clinician PICU-QODD.

CONTEXT: In the pediatric intensive care setting, an accurate measure of the dying and death experience holds promise for illuminating how critical care nurses, physicians, and allied psychosocial staff can better manage end-of-life care for the benefit of children and their families, as well as the caregivers. OBJECTIVES: The aim was to assess the reliability and validity of a clinician measure of the quality of dying and death (Pediatric Intensive Care Unit-Quality of Dying and Death 20 [PICU-QODD-20]) in the pediatric intensive care setting. METHODS: In a retrospective cohort study, five types of clinicians (primary nurse, bedside nurse, attending physician, and the psychosocial clinician and critical care fellow most involved in the case) were asked to complete a survey for each of the 94 children who died over a 12 month period in the pediatric intensive care units of two children's hospitals in the northeast U.S. Analyses were conducted within type of clinician. RESULTS: In total, 300 surveys were completed by 159 clinicians. Standard item analyses and substantive review led to the selection of 20 items for inclusion in the PICU-QODD-20. Cronbach alpha for the PICU-QODD-20 ranged from 0.891 for bedside nurses to 0.959 for attending physicians. For each type of clinician, the PICU-QODD-20 was significantly correlated with the quality of end-of-life care and with meeting the family's needs. In addition, when patient/family or team barriers were encountered, the PICU-QODD-20 score tended to be significantly lower than for cases in which the barrier was not encountered. CONCLUSION: The PICU-QODD-20 shows promise as a valid and reliable measure of the quality of dying and death in pediatric intensive care.

Desipramine enhances the ability of risperidone to decrease alcohol intake in the Syrian golden hamster.

The atypical antipsychotic clozapine reduces alcohol drinking in patients with schizophrenia. We have proposed that clozapine׳s ability to decrease alcohol drinking relates to its weak blockade of the dopamine D2 receptor and potent blockade of the norepinephrine α-2 receptor, as well as its ability to elevate plasma and brain norepinephrine. Another atypical antipsychotic, risperidone, which is a potent blocker of both the dopamine D2 receptor and norepinephrine α-2 receptor, does not decrease alcohol drinking. In this study, we used the Syrian golden hamster to test whether the ability of risperidone to reduce alcohol drinking would be enhanced if it was used in combination with the norepinephrine reuptake inhibitor desipramine. Hamsters were given free access to water and alcohol (15% v/v) until they reached a steady drinking baseline. They were then treated daily with each drug or drug combination for 20 days. Risperidone (0.2mg/kg) only transiently decreased alcohol drinking. However, 5.0mg/kg, and possibly 1.0mg/kg, desipramine added to 0.2mg/kg risperidone appeared to produce a more substantial and relatively sustained effect than risperidone alone. Data from this study provide leads toward the development of new treatments for patients with schizophrenia and alcoholism, and also for those with alcoholism alone.

Introduction to dynamic treatment strategies and sequential multiple assignment randomization.

BACKGROUND: In June 2013, a 1-day workshop on Dynamic Treatment Strategies (DTSs) and Sequential Multiple Assignment Randomized Trials (SMARTs) was held at the University of Pennsylvania in Philadelphia, Pennsylvania. These two linked topics have generated a great deal of interest as researchers have recognized the importance of comparing entire strategies for managing chronic disease. A number of articles emerged from that workshop. PURPOSE: The purpose of this survey of the DTS/SMART methodology (which is taken from the introductory talk in the workshop) is to provide the reader the collected articles presented in this volume with sufficient background to appreciate the more detailed discussions in the articles. METHODS: The way that the DTS arises naturally in clinical practice is described, along with its connection to the well-known difficulties of interpreting the analysis by intention-to-treat. The SMART methodology for comparing DTS is described, and the basics of estimation and inference presented. RESULTS: The DTS/SMART methodology can be a flexible and practical way to optimize ongoing clinical decision making, providing evidence (based on randomization) for comparative effectiveness. LIMITATIONS: The DTS/SMART methodology is not a solution for unstandardized study protocols. CONCLUSIONS: The DTS/SMART methodology has growing relevance to comparative effectiveness research and the needs of the learning healthcare system.

The comparative effects of clozapine versus haloperidol on initiation and maintenance of alcohol drinking in male alcohol-preferring P rat.

Alcohol use disorder, characterized by modest levels of alcohol use, commonly occurs in patients with schizophrenia and dramatically worsens their course. Recent data indicate that the atypical antipsychotic clozapine, but not the typical antipsychotic haloperidol, decreases alcohol drinking both in patients with schizophrenia and also in the Syrian golden hamster, an animal model of moderate alcohol drinking. The present study was designed to assess the comparative effects of clozapine and haloperidol in the alcohol-preferring (P) rat, an animal model of alcoholism. First, the study investigated the comparative effects of clozapine and haloperidol on initiation of alcohol consumption in P rats, which models the early stage of alcoholism. Second, the study assessed the comparative effects of clozapine and haloperidol on maintenance of chronic alcohol consumption in P rats to provide a clue as to whether either drug may also limit alcohol consumption in alcohol-dependent patients. Clozapine attenuated the initiation of alcohol drinking and development of alcohol preference while haloperidol did not. However, neither clozapine nor haloperidol attenuated maintenance of chronic alcohol drinking. Taken together, the current data suggest that clozapine, but not haloperidol, may be effective at reducing alcohol abuse or non-dependent drinking and the P rat, used within an alcohol initiation paradigm, and may differentiate the effects of clozapine and haloperidol on alcohol drinking.

Longitudinal trends in health-related quality of life in adults with cystic fibrosis.

BACKGROUND: Health-related quality-of-life (HRQOL) measures have been used as patient-reported outcomes in clinical trials in cystic fibrosis (CF), but there are limited data on HRQOL changes over time in adults with CF. METHODS: The Project on Adult Care in Cystic Fibrosis, a prospective, longitudinal panel study of 333 adults with CF at 10 CF centers in the United States, administered a disease-specific HRQOL measure, the Cystic Fibrosis Questionnaire-Revised (CFQ-R), seven times over 21 months. The CFQ-R assesses both physical and psychosocial domains of health. Growth curve regression models were developed for each CFQ-R domain, adjusting for demographic and clinical characteristics. RESULTS: Between 205 and 303 adults completed surveys (response rate, 70%-93%). Mean age at baseline was 33 years (range, 19-64 years); mean FEV1 % predicted was 59.8% (SD, 22%). Over the 21 months of follow-up, lung function, frequency of pulmonary exacerbations, and nutritional indices were associated with physical CFQ-R domain scores. There were no significant population trends over time in the physical domain scores; however, there were population time trends in three psychosocial domains: treatment burden (+8.9 points/y), emotional functioning (+3.2 points/y), and social functioning (-2.4 points/y). Individual variation in both physical and psychosocial subscales was seen over 21 months. CONCLUSIONS: In a longitudinal multicenter population of adults with CF, clinical variables such as FEV1, exacerbation frequency, and weight were correlated with related CFQ-R subscales. For the population as a whole, the physical domains of CFQ-R, such as respiratory symptoms, were stable. In contrast, population changes in several psychosocial domains of CFQ-R suggest that differentiating between the physical and the psychosocial trajectories in health among adults with CF is critical in evaluating patient-reported outcomes.

Efficient design and inference for multistage randomized trials of individualized treatment policies.

Clinical demand for individualized "adaptive" treatment policies in diverse fields has spawned development of clinical trial methodology for their experimental evaluation via multistage designs, building upon methods intended for the analysis of naturalistically observed strategies. Because often there is no need to parametrically smooth multistage trial data (in contrast to observational data for adaptive strategies), it is possible to establish direct connections among different methodological approaches. We show by algebraic proof that the maximum likelihood (ML) and optimal semiparametric (SP) estimators of the population mean of the outcome of a treatment policy and its standard error are equal under certain experimental conditions. This result is used to develop a unified and efficient approach to design and inference for multistage trials of policies that adapt treatment according to discrete responses. We derive a sample size formula expressed in terms of a parametric version of the optimal SP population variance. Nonparametric (sample-based) ML estimation performed well in simulation studies, in terms of achieved power, for scenarios most likely to occur in real studies, even though sample sizes were based on the parametric formula. ML outperformed the SP estimator; differences in achieved power predominately reflected differences in their estimates of the population mean (rather than estimated standard errors). Neither methodology could mitigate the potential for overestimated sample sizes when strong nonlinearity was purposely simulated for certain discrete outcomes; however, such departures from linearity may not be an issue for many clinical contexts that make evaluation of competitive treatment policies meaningful.

Raclopride lessens the ability of clozapine to suppress alcohol drinking in Syrian golden hamsters.

Emerging evidence suggests that the atypical antipsychotic clozapine decreases alcohol consumption in patients with schizophrenia, while typical antipsychotics, all of which are potent dopamine (DA) D2 receptor antagonists, do not. We have proposed that clozapine, through its weak DA D2 receptor blocking action, coupled with its ability to potentiate noradrenergic and serotonergic activity, may ameliorate a dysfunction in the mesocorticolimbic DA reward circuitry that underlies alcohol use disorder in patients with schizophrenia. In prior studies, we have demonstrated that clozapine also decreases alcohol drinking in the Syrian golden hamster, but haloperidol does not. The purposes of the current study were: (1) to further assess the effect of clozapine (2 or 4 mg/kg/day, s.c.) on alcohol consumption in hamsters, using a continuous access, 2-bottle choice paradigm; and (2) to examine whether clozapine's effect on alcohol drinking is affected by increasing its DA D2 blockade through adjunctive use of the potent DA D2 receptor antagonist raclopride (2, 4, or 6 mg/kg/day, s.c.). The major findings were: (1) clozapine suppressed both initiation and maintenance of alcohol drinking in hamsters; and (2) these effects of clozapine were lessened when raclopride was given adjunctively with clozapine. These data suggest that clozapine may limit alcohol drinking in the golden hamster (and possibly in patients with schizophrenia) in part because of its weak blockade of the DA D2 receptor.

A randomized trial of clozapine vs. other antipsychotics for cannabis use disorder in patients with schizophrenia.

OBJECTIVE: Cannabis use disorder is the most common co-occurring drug use disorder in people with schizophrenia and is associated with poor outcomes. We launched a randomized controlled trial to assess the impact of clozapine compared with treatment as usual on cannabis use in patients with schizophrenia and co-occurring cannabis use disorder. METHODS: Thirty-one patients with schizophrenia and co-occurring cannabis use disorder were randomly assigned to switch to clozapine or to stay on their current antipsychotic and were then followed weekly for 12 weeks. Blinded raters assessed participants weekly with the Timeline Follow-back for substance use and the expanded Brief Psychiatric Rating Scale for symptoms. Longitudinal random effects models were used to investigate the time-varying differences in cannabis use and other outcomes between the treatment as usual and clozapine groups. RESULTS: The two groups differed in average intensity of cannabis use by approximately 4.5 joints/week, with lesser use in the clozapine group (t = -1.77; df = 28.5; p=.086; effect size ~ 0.6). Symptoms and functioning were not different between the two groups. CONCLUSIONS: Clozapine may reduce cannabis use among patients with schizophrenia and co-occurring cannabis use disorder. Further controlled trials are warranted.

Sample size calculations for evaluating treatment policies in multi-stage designs.

BACKGROUND: Sequential multiple assignment randomized (SMAR) designs are used to evaluate treatment policies, also known as adaptive treatment strategies (ATS). The determination of SMAR sample sizes is challenging because of the sequential and adaptive nature of ATS, and the multi-stage randomized assignment used to evaluate them. PURPOSE: We derive sample size formulae appropriate for the nested structure of successive SMAR randomizations. This nesting gives rise to ATS that have overlapping data, and hence between-strategy covariance. We focus on the case when covariance is substantial enough to reduce sample size through improved inferential efficiency. METHODS: Our design calculations draw upon two distinct methodologies for SMAR trials, using the equality of the optimal semi-parametric and Bayesian predictive estimators of standard error. This 'hybrid' approach produces a generalization of the t-test power calculation that is carried out in terms of effect size and regression quantities familiar to the trialist. RESULTS: Simulation studies support the reasonableness of underlying assumptions as well as the adequacy of the approximation to between-strategy covariance when it is substantial. Investigation of the sensitivity of formulae to misspecification shows that the greatest influence is due to changes in effect size, which is an a priori clinical judgment on the part of the trialist. LIMITATIONS: We have restricted simulation investigation to SMAR studies of two and three stages, although the methods are fully general in that they apply to 'K-stage' trials. CONCLUSIONS: Practical guidance is needed to allow the trialist to size a SMAR design using the derived methods. To this end, we define ATS to be 'distinct' when they differ by at least the (minimal) size of effect deemed to be clinically relevant. Simulation results suggest that the number of subjects needed to distinguish distinct strategies will be significantly reduced by adjustment for covariance only when small effects are of interest.

Clozapine chronically suppresses alcohol drinking in Syrian golden hamsters.

Alcohol use disorder is common in patients with schizophrenia and is associated with poor clinical outcomes. Preliminary reports from our group and others suggest that the atypical antipsychotic clozapine may decrease alcohol use in these patients. We have previously shown that clozapine suppresses alcohol consumption for 9 days in Syrian golden hamsters. Here, we assessed the effects of clozapine on alcohol consumption in hamsters over a 27-day period, using a continuous access, 2-bottle (15% alcohol vs. water) protocol. Clozapine (4, 8, or 12 mg/kg/day, injected subcutaneously [s.c.]) dose-dependently suppressed alcohol drinking, while increasing food and water intake. There was no tolerance within individual groups to the effect of clozapine on alcohol drinking over time. In a separate experiment, the effects of clozapine on sucrose and water drinking and food intake over a 9-day period were assessed. Clozapine (4, 8, or 12 mg/kg/day s.c.) failed to suppress sucrose (0.09 M), food, or water consumption at any time-point tested. In a related study, assessment of blood alcohol levels in hamsters indicated that blood alcohol levels were maintained within a narrow and moderate range (7-13 mg/dL), levels noted by others to produce physiologic effects in rodents. The ability of clozapine to suppress alcohol drinking in the hamster over an extended period of time without suppressing sucrose, water, or food consumption is consistent with preliminary reports indicating that clozapine limits frequent alcohol use, even producing abstinence in many patients with schizophrenia.

Developing and testing adaptive treatment strategies using substance-induced psychosis as an example.

Decisions concerning treatment changes pervade the management of chronic psychiatric disorders that resist definitive cure, yet empirical evidence for the comparative clinical effectiveness of treatment strategies remains underdeveloped. In this paper we exploit the example of psychosis following substance use to illustrate some new developments in clinical trials design that can provide the most solid evidence base for defining successful strategies. The intent is to explore the strengths and limitations of the methodological approach through a meaningful clinical example, with an emphasis on concepts and issues. Both methodology and clinical science are overviewed.

Sequential causal inference: application to randomized trials of adaptive treatment strategies.

Clinical trials that randomize subjects to decision algorithms, which adapt treatments over time according to individual response, have gained considerable interest as investigators seek designs that directly inform clinical decision making. We consider designs in which subjects are randomized sequentially at decision points, among adaptive treatment options under evaluation. We present a sequential method to estimate the comparative effects of the randomized adaptive treatments, which are formalized as adaptive treatment strategies. Our causal estimators are derived using Bayesian predictive inference. We use analytical and empirical calculations to compare the predictive estimators to (i) the 'standard' approach that allocates the sequentially obtained data to separate strategy-specific groups as would arise from randomizing subjects at baseline; (ii) the semi-parametric approach of marginal mean models that, under appropriate experimental conditions, provides the same sequential estimator of causal differences as the proposed approach. Simulation studies demonstrate that sequential causal inference offers substantial efficiency gains over the standard approach to comparing treatments, because the predictive estimators can take advantage of the monotone structure of shared data among adaptive strategies. We further demonstrate that the semi-parametric asymptotic variances, which are marginal 'one-step' estimators, may exhibit significant bias, in contrast to the predictive variances. We show that the conditions under which the sequential method is attractive relative to the other two approaches are those most likely to occur in real studies.

Adaptive treatment strategies in chronic disease.

An adaptive treatment strategy (ATS) is a rule for adapting a treatment plan to a patient's history of previous treatments and the response to those treatments. The ongoing management of chronic disease defines an ATS, which may be implicit and hidden or explicit and well-specified. The ATS is characterized by the use of intermediate, early markers of response to dynamically alter treatment decisions, in order to achieve a favorable ultimate outcome. We illustrate the ATS concept and describe how the effect of initial treatment decisions depends on the performance of subsequent decisions at later stages. We show how to compare two or more ATSs, or to determine an optimal ATS, using a sequential multiple assignment randomized (SMAR) trial. Designers of clinical trials might find the ATS concept useful in improving the efficiency and ecological relevance of clinical trials.

Improving the efficiency of estimation in randomized trials of adaptive treatment strategies.

BACKGROUND: Given the history of treatments to date, and the responses of the patient, what is the best treatment to try next? An ensemble of sequential, multistage rules guiding such adaptive decision making can be described as an ;adaptive treatment strategy (ATS)'. Robins' G-computation can be used for estimation of the mean outcome of an ATS from a ;sequential multiple assignment randomized (SMAR)' trial. PURPOSE: To develop a variance estimate for the G-computation formula, based on a sequential analysis of the states and treatments observed in the trial, and compare its properties with those of the ;marginal mean' method described by Murphy, which is based on an estimating equation. METHODS: We use both mathematical calculation and simulation studies to demonstrate the properties of the G-computation and its sequential variance estimate, including finite-sample bias and coverage. RESULTS: The sequential method is unbiased and more efficient when the variation in intervening states contributes substantially to the variation in final outcome, and when the study can be designed to guarantee full observation of the ATS under study. The method extends to the comparison of two or more ATS. LIMITATIONS: If full observation cannot be guaranteed, the method may have poor finite-sample properties. CONCLUSIONS: When the states used to adapt treatment contribute substantially to the outcome, and good design technique can be applied, the sequential method provides more efficient estimation.

Adaptive strategies for treating childhood mania.

The conduct of trials in children is beset with special difficulties associated with the dearth of treatment data, the considerable heterogeneity of pediatric patient populations connected with (for example) developmental stage, and the strong desire of parents to see that children are provided with the best possible care at all times. To address these issues, we propose the adaptive treatment strategy (ATS) study in which medication changes are adaptively determined according to the evolving treatment response of the individual child. To formalize this methodological approach, we parameterize an ATS as a "threshold" decision rule that monitors whether the patient's response trajectory crosses some threshold of failure. In this formulation, the threshold represents a priori judgments about when to give up on response to medication, and the goal is to find the right response thresholds for continuation, augmentation, or switching. Our exposition is developed in the specific clinical context of childhood mania to maximize accessibility of the ideas but applies more generally to other chronic mental and physical health disorders that are difficult to treat.