Search Results for: smart

Building Better Adaptive Interventions by Expanding SMART

June 27, 2019:

John DziakBehavioral interventions for prevention and treatment are an important part of the fight against drug abuse and HIV/AIDS. Among the challenges faced by scientists is how and when to alter the course of treatment for participants in the intervention. Adaptive interventions change based on evidence about what is best for the participant at a given time.

For over a decade, Methodology Center researchers have developed and applied sequential, multiple assignment, randomized trials (SMARTs), which are experimental designs that can be used to build adaptive interventions that address a variety of health and behavioral challenges, such as substance abuse abstinence, weight loss, ADHD management, and language acquisition. Recently, researchers have begun developing methods to evaluate SMARTs by using multiple measures of the outcome over time rather than only considering the outcome at the end of the study. For example, a researcher who is developing an adaptive intervention to promote abstinence from alcohol may want to consider alcohol usage rates every month for six months to decide how to construct the intervention. In a recent article in Multivariate Behavioral Research by Methodology Center Investigator John Dziak, Methodology Center Affiliates Daniel Almirall and Inbal “Billie” Nahum-Shani, and others, the authors develop and demonstrate a new method for evaluating a SMART using repeated measures of a binary outcome (such as substance use versus nonuse).

The authors apply their method to the ENGAGE SMART study, which was conducted to help develop an adaptive intervention for promoting treatment engagement among cocaine- and alcohol-dependent individuals. The authors found that certain designs correlated to increased abstinence rates during the first two months but abstinence rates that were equivalent to other designs by the end of the study. Had the investigators measured relapse solely at six months, they would not have observed the relapse differences during the early months, which may have practical or clinical significance. The authors go on to provide guidelines for using multiple binary measurements of the outcome while analyzing data from a SMART.

Lead author John Dziak discussed the importance of the study. “SMART is a valuable method because conditions such as addiction and many other health problems, are chronic and often need treatment over time. In many cases, the appropriate treatment could change depending on the individual’s experiences. SMART trials can help scientists decide which set of adaptive treatment rules will work the best. In a lot of the past SMART literature, ‘work the best’ just meant having the best expected outcome at the end of the study.  But considering short-term and long-term effects together might help clinicians make better decisions to fit an individual’s  goals.  Also, it allows scientists to study delayed effects, where an early treatment choice affects how well later treatments work, and that could render theoretical insight into the treatments.”

Reference

Dziak, J. J., Yap, J. R., Almirall, D., McKay, J. R., Lynch, K. G., & Nahum-Shani, I. (2019). A data analysis method for using longitudinal binary outcome data from a SMART to compare adaptive interventions. Multivariate Behavioral Research, 1-24.

Pilot SMART for Elementary Students With Autism Spectrum Disorder

This project will develop an adaptive intervention to improve social connectedness, academic engagement, and other skills among school-aged children with autism spectrum disorder. Treatment for each participant includes some combination of a playground-based intervention, a classroom-based intervention, a peer-mediated intervention, and a parent-assisted intervention.​ This pilot project will address feasibility and acceptability concerns and will provide preliminary data for a full-scale SMART.

  • PI: Connie Kasari
  • Location: Center for Autism Research and Treatment, University of California, Los Angeles
  • Funding: Funded by Institute of Educational Sciences

Pilot SMART for Treating ADHD in Families

This project aims to develop an adaptive intervention for families where the mother has ADHD and the child is at genetic and environmental risk for ADHD. Researchers are using SMART to determine how to use behavioral training or medication for mothers separately, in sequence, or in combination, to improve the quality of parenting and prevent the progression of ADHD in children.
  • PIs: Mark Stein and Andrea Chronis-Tuscano
  • Location: Seattle Children’s Hospital
  • Funding: NIMH Project R34MH99208

SMART Design for Attendance-Based Prize Contingency Management (CM) for Cocaine Abuse

Contingency management (CM) is a treatment used in substance abuse where patients are rewarded for following treatment guidelines. In this study, researchers are comparing CM to treatment without incentives using a SMART design. They are also testing the timing and the length of the CM.
  • PI: Nancy Petry
  • Location: University of Connecticut Health Center
  • Funding: NIDA Project R01AA021446
 

Pilot SMART for Personalizing Treatment for Child Depression

This pilot project uses a SMART design to develop an adaptive intervention for children with depression. Dr. Eckshtain aims to develop decision rules regarding the use of cognitive behavioral treatment, caregiver–child treatment, or both. The goal is to develop an adaptive treatment protocol for depressed children.
  • PI: Dikla Eckshtain
  • Location: Judge Baker Children’s Center at Harvard Medical School
  • Funding: NIMH Project K23MH093491

Pilot SMART for Adolescent Depression

Adolescents suffering from depression begin treatment with interpersonal psychotherapy. This pilot project employs SMART to help establish treatment rules regarding when and in what way to intensify treatment.
  • PI: Meredith L. Gunlicks-Stoessel
  • Location: University of Minnesota, Twin Cities
  • Funding: NIMH Project K23MH090216

Projects Using SMARTs

This page contains several examples of SMART designs in order to illustrate the different design possibilities and questions that SMART can answer. Click on an image to enlarge and show detail. To see a subset of the designs that relate to a specific health problem or design type, select from the dropdown list.

Last updated in 2016.

Adaptive Treatment for Pregnant Women Who Abuse Drugs

Researchers have developed an intensive relapse-prevention program for pregnant women who abuse drugs. A SMART design is being used to develop an adaptive intervention where the intensity and scope of the relapse-prevention program is adjusted based on the evolving status of the woman.

PIs: Hendrée Jones, Margaret Chisolm
Location: Johns Hopkins University
Funding: NIDA Project R01DA014979

See a detailed version of the schematic.

 

 

Adaptive Treatment for Growth Suppression in Children with ADHD

Studies show that the use of stimulants for the control of ADHD in youth leads to a reduction in height gain. This study uses a SMART design to examine the effectiveness of temporary breaks in medicinal treatments and caloric supplementation for the treatment of stimulant-induced weight and growth suppression.

PI: James G. Waxmonsky
Location: Florida International University
Funding: NIMH Project R01MH083692

See a more detailed version of the schematic.

 

Adaptive Intervention for Adolescent Marijuana Use

Researchers in this study are developing an adaptive treatment for adolescent marijuana users. They are studying the use and combination of several efficacious treatments, including behavioral therapy, contingency management, behavioral parent training, and working memory training via a SMART trial.

PI: Alan J. Budney
Location: Dartmouth College
Funding: NIDA Project R01DA015186

See a detailed version of the schematic.

 

 

 

Adaptive Interventions for Children with ADHD

The aim of this SMART is to understand whether to begin with medication or behavioral therapy for children with ADHD, and whether to intensify or augment initial treatment for children who do not respond to treatment.

PI: William Pelham
Location: Florida International University
Funding: U.S. Department of Education-funded, completed project

See a detailed version of the schematic.

 

 

 

Adaptive Treatment for Cocaine Dependence

A SMART design is being implemented to develop an adaptive intervention to increase treatment engagement and decrease cocaine use for patients who are cocaine dependent. The study also examines whether patient choice of care affects patient outcomes.

PI: James R. McKay
Location: University of Pennsylvania
Funding: NIDA Project P01AA016821

See a detailed version of the schematic.

 

 

Adaptive Approach to Naltrexone Treatment for Alcoholism

Naltrexone (NTX) is an opioid receptor antagonist used to prevent alcoholism relapse. This trial examines how to define “non-response” to treatment with NTX and what treatments are most effective for those who do or do not respond to the initial treatment.

PI: David Oslin
Location: University of Pennsylvania
Funding: NIAAA Project R01AA017164

See a detailed version of the schematic.

 

 

Adaptive Treatment for Adolescent Obesity

This project targets African American adolescents with obesity and their parents. SMARTs are used to develop an adaptive intervention that increases skills in changing dietary, exercise, and sedentary behaviors.

PI: Sylvie Naar-King
Location: Wayne State University
Funding: NHLBI Project U01HL097889

See a detailed version of the schematic.

 

 

Adaptive Treatment Strategies for Children and Adolescents With Obsessive-Compulsive Disorder (OCD)

For youth with OCD, the most common treatments are cognitive-behavioral therapy (CBT), pharmacological treatment, or both. Up to 30% of patients may not benefit from their initial treatments. Researchers will employ a SMART to determine the optimal treatment sequence for participants dependent on whether or not they respond to their initial treatment.

PI: Roseli Shavitt
Location: University of Sao Paulo

See a detailed version of the schematic.

 

 

Pilot SMART for Adolescent Depression

Adolescents suffering from depression begin treatment with interpersonal psychotherapy. This pilot project employs SMART to help establish treatment rules regarding when and in what way to intensify treatment.

PI: Meredith L. Gunlicks-Stoessel
Location: University of Minnesota, Twin Cities
Funding: NIMH Project K23MH090216 

See a detailed version of the schematic.

 

 

 

This project aims to develop an adaptive intervention for persistent insomnia. Researchers are using SMART to determine the best sequencing of cognitive behavioral therapy and medication for persistent insomnia.

PI: Charles Morin
Location: Laval University
Funding: NIMH Project R01MH091053

See a detailed version of the schematic.

 

 

Patients suffering from bipolar disorder are assigned to one of two mood stabilizers. A SMART design is used to determine the appropriate treatment for patients who develop depression.

PIs: Charles Lee Bowden, Joseph Calabres
Locations: University of Texas Health Science Center at San Antonio, 2nd site: Case Western Reserve University Medical Center
Funding: NIMH Project P30MH08604

See a detailed version of the schematic.

 

 

This pilot project uses a SMART design to develop an adaptive intervention for children with depression. Dr. Eckshtain aims to develop decision rules regarding the use of cognitive behavioral treatment, caregiver–child treatment, or both. The goal is to develop an adaptive treatment protocol for depressed children.

PI: Dikla Eckshtain
Location: Judge Baker Children’s Center at Harvard Medical School
Funding: NIMH Project K23MH093491

See a detailed version of the schematic.

 

 

Characterizing Cognition in Nonverbal Individuals With Autism

In order to develop communication skills among school-aged children who are nonverbal, this project employs a SMART design to test a novel intervention. The intervention includes components that focus on spoken language and the use of a speech-generating device (e.g., iPad). The SMART design provides the data needed to define response and nonresponse to the intervention and identify the best treatment sequence.

PI: Connie Kasari
Location: Center for Autism Research and Treatment, University of California, Los Angeles
Funding: Funded by Autism Speaks

See a detailed version of the schematic.

 

Adaptive Interventions for Minimally Verbal Children With Autism Spectrum Disorder in the Community

This study will compare two types of intensive, daily instruction for children with autism spectrum disorder (ASD) who use only minimal verbal communication. Earlier research has shown that even after early language-skills training, about one-third of school-aged children with ASD remain minimally verbal. Researchers plan to enroll 200 children in four cities: Los Angeles, Nashville, New York City, and Rochester, N.Y.

PI: Connie Kasari
Location: Center for Autism Research and Treatment, University of California, Los Angeles
Funding: NICHD Project R01HD073975

See a detailed version of the schematic.

 

Adaptive Intervention Strategies in Conduct Problem Prevention: Pilot Study

This study compares two types of interventions for youth (ages 10-15) with conduct disorders. Participants received either a teen-focused or parent-focused intervention. The appropriate intensity of the interventions was also studied.

PI: Gerald August
Location: University of Minnesota
Funding: NIMH Project R34MH097832

See a more detailed version of the schematic.

 

 

SMART Design for Attendance-Based Prize Contingency Management (CM) for Cocaine Abuse

Contingency management (CM) is a treatment used in substance abuse where patients are rewarded for following treatment guidelines. In this study, researchers are comparing CM to treatment without incentives using a SMART design. They are also testing the timing and the length of the CM.

PI: Nancy Petry
Location: University of Connecticut Health Center
Funding: NIDA Project R01AA021446

See a detailed version of the schematic.

 

 

Adaptive Treatment for Smoking Among People With HIV

Between 50% and 70% of people living with HIV are nicotine dependent. This SMART examines how and when to apply contingency management and standard treatment to promote smoking cessation in this population.

PI: David Ledgerwood
Location: Wayne State
Funding: NIMH Project R01DA034537

See a detailed version of the schematic.

 

 

 

Pilot SMART for Treating ADHD in Families

This project aims to develop an adaptive intervention for families where the mother has ADHD and the child is at genetic and environmental risk for ADHD. Researchers are using SMART to determine how to use behavioral training or medication for mothers separately, in sequence, or in combination, to improve the quality of parenting and prevent the progression of ADHD in children.

PIs: Mark Stein and Andrea Chronis-Tuscano
Location: Seattle Children’s Hospital
Funding: NIMH Project R34MH99208

See a detailed version of the schematic.

 

 

Improving Mental Health Outcomes: Building an Adaptive Implementation Strategy

This SMART is cluster-randomized. Randomization occurs at the clinic level. The aim of the study is to develop an adaptive quality improvement strategy designed to enhance the implementation of an evidence-based mental health intervention. Outcomes are measured at the patient level. 

PI: Amy Kilbourne
Location: University of Michigan
Funding: NIMH Project R01MH099898

See a detailed version of the schematic.

 

 

Adaptive Intervention for Suicide Prevention Among College Students

Researchers in this study are developing an adaptive treatment to address suicidality in college students seeking services at college counseling centers. They are developing the first empirically validated approach to sequence treatments for students seeking services.

PI: Jacqueline Pistorello
Location: University of Nevada, Reno
Funding: NIMH Project R34MH104714

 

 

Pilot SMART for Elementary Students With Autism Spectrum Disorder

This project will develop an adaptive intervention to improve social connectedness, academic engagement, and other skills among school-aged children with autism spectrum disorder. Treatment for each participant includes some combination of a playground-based intervention, a classroom-based intervention, a peer-mediated intervention, and a parent-assisted intervention.​ This pilot project will address feasibility and acceptability concerns and will provide preliminary data for a full-scale SMART.

PI: Connie Kasari
Location: Center for Autism Research and Treatment, University of California, Los Angeles
Funding: Funded by Institute of Educational Sciences 

See a detailed version of the schematic.

 

 

Adolescent Substance Abuse: Progressive Treatment for Adolescent Who Use Drugs

Because the history of adolescent substance abuse interventions shows that individuals respond differently to treatment, this study uses a pair of SMART designs to examine when and how to treat adolescent drug users.

PI: Holly Barrett Waldron
Location: Oregon Research Institute
Funding: NIDA-funded, completed project

 

Adaptive Intervention Developed Using SMART

Introductory Example: Using Medication to Prevent Alcoholism Relapse

a glass of wine in front of a despondent-looking woman in the backgroundThe SMART in our example is the Extending Treatment Effectiveness of Naltrexone (ExTENd) trial of alcohol-dependence treatments led by David Oslin. In this study, researchers studied the effectiveness of Naltrexone (NTX), an opioid receptor antagonist. An opioid receptor antagonist is a drug that bonds to the opioid receptors in the brain without provoking a response from the receptors. NTX has been used to prevent relapse to alcoholism, but because it diminishes the pleasurable effects of drinking, participants often fail to adhere to NTX regimens.

The researchers sought to answer the following questions:

  • What level of drinking should be used to define “nonresponse” to NTX?
  • What treatment should be used when participants do not respond to NTX?
  • What treatment should be used to prevent relapse among individuals who responded to NTX?

The SMART design tested eight different adaptive interventions that were candidates for how to best use NTX.

 

What is a SMART design?

Adaptive Intervention:
An intervention that adapts the type or dosage of a treatment based on patient characteristics or response

Adaptive interventions adapt the type or dosage of the intervention based on patient characteristics. Then, the treatment is adjusted repeatedly over time. Because this is a multi-stage process, an adaptive intervention can use a series of decision rules about when and how to modify the intervention. These interventions use individual differences between participants to achieve the best possible outcome, whether that means augmenting an intervention for a non-responsive participant or diminishing treatment for a responsive participant in order to reduce cost or participant burden.

A SMART provides high-quality data for the construction of adaptive interventions. In a SMART, the decision rules for an adaptive intervention are tested. Each participant is randomized into different treatment options when they reach decision points throughout the intervention. A SMART is not an adaptive intervention, but rather a trial that contains multiple adaptive interventions. In a SMART, treatments for participants are randomized at each stage, and the data from the study is then used to design an adaptive intervention in which participants are not randomized: their treatments change based on the intervention’s decision rules.

The example below provides details about SMART designs.

 

About The Study

Example: SMART design for alcohol-dependence intervention

First, all participants were randomized. Both conditions received NTX and MM, but one group was classified as responsive/ nonresponsive using the stringent definition, and the other using the lenient definition. As soon as a participant met the criteria for non-response, s/he was re-randomized. After eight weeks, if a participant had not met the criteria for non-response, s/he was classified as responsive. Participants who were non-responsive to NTX+MM were assigned to CBI + MM and were randomized to receive either NTX or a placebo. Participants who responded to initial NTX+MM were assigned either to NTX with no additional support, or to NTX + Phone support. In total, the study lasted 6 months.

The graphic represents the SMART design used in the ExTENd study.

Random = a randomization of participants.
NTX = naltrexone
MM = medical management (face-to-face, basic, minimal clinical support for NTX – or placebo – use)
CBI = combined behavioral intervention (more intensive, multicomponent intervention. Targets adherence to NTX and motivation.)
Phone = telephone disease management (like MM, but delivered via telephone)
Stringent = 2+ days of heavy drinking is defined as non-responsive to NTX.
Lenient =  5+ days of heavy drinking is defined as non-responsive to NTX.

For more information about SMART designs and this study, see

Lei, H., Nahum-Shani, I., Lynch, K., Oslin, D., & Murphy, S. A. (2012). A “SMART” design for building individualized treatment sequences. Annual Review of Clinical Psychology, 8, 14.1 – 14.28. doi: 10.1146/annurev-clinpsy-032511-143152​

 

 

 

Data

SMART design:
a trial that provides data for the construction of adaptive interventions

This study had a usable sample size of 250. Multiple imputation was used in the analyses of data for participants who dropped out during the second phase of treatment.

  • 70% white
  • 86% male
  • 28% over age 55

Analysis and results

Lei et al (2012) performed an illustrative analysis on the data. The authors reveal how SMART can provide valuable information about the tailoring variable and indicate which treatments were most efficient and effective.

References

Lei, H., Nahum-Shani, I., Lynch, K., Oslin, D., & Murphy, S. A. (2012). A “SMART” design for building individualized treatment sequences. Annual Review of Clinical Psychology, 8, 14.1 – 14.28. doi: 10.1146/annurev-clinpsy-032511-143152

Nahum-Shani, I., Ertefaie, A., Lu, X., McKay, J.R., Lynch, K.G., Oslin, D., & Almirall, D. (2017). A SMART Data Analysis Method for Constructing Adaptive Treatment Strategies for Substance Use Disorders. Addiction, 112(5), 901-909.

Read the rationale for using SMART designs.

See our recommended reading for SMART.

 

Last updated: May 11, 2020

Why Use a SMART Design to Build an Adaptive Intervention?

Adaptive Interventions

blank SMART diagramAdaptive interventions have four critical components.

  1. Sequence of decisions regarding patient care – Most interventions require decisions such as, “If the patient is unresponsive to the initial treatment, what treatment should we provide next?” or “Once the patient has stabilized, what treatment is needed to prevent relapse?”
  2. The set of treatment options at each decision point – For example, if a patient is unresponsive to a drug, should the dosage be increased, should the drug be discontinued, or should counseling be increased? All of these are treatment options.
  3. Tailoring variables – These are the factors used to trigger a change in the treatment. These can be things like early signs of nonresponse, manifestation of side effects, or environmental or social characteristics. The idea is to identify the variables that best indicate when the appropriate treatment has changed.
  4. A sequence of decision rules – This links the first three components. There should be one decision rule per decision. The tailoring variables provide information about which of the treatment options is most appropriate for the patient at the time of the decision.
A series of decision rules must be established to guide treatment based on individual characteristics.

Adaptive interventions allow researchers

  • to help patients who do not respond to initial treatment,
  • to respond if the effectiveness of an intervention wanes over time due to changes in the patient’s situation or response,
  • to prioritize when the patient possesses comorbid conditions (e.g., depression and alcoholism),
  • to address relapses (as are common when treating substance use),
  • to decrease burden and/or cost of the intervention when a patient is stable, and
  • to respond when patients do not adhere to a treatment.

Why Conduct a SMART?

SMART designs allow for the testing of multiple potential adaptive interventions along with the proposed tailoring variables.

Sequential, multiple assignment, randomized trials (SMARTs) provide data that enables the development of high-quality adaptive interventions. In a SMART there is a separate stage for each of the critical decisions involved in the adaptive intervention. At each stage, all participants are randomly assigned to a treatment option. By randomizing participants multiple times, scientists can assess the effectiveness of each stage. So, several adaptive interventions are embedded within each SMART design for testing. This allows testing of the tailoring variables and the intervention components in the same trial, and it allows clinicians to develop the best decision rules based on research rather than a priori decisions. Read about a SMART that tested the level of the tailoring variable and the treatment options for the construction of an adaptive intervention for relapsing alcoholics.

For more information about SMART designs see Lei et al. (2012).

 

SMART vs Other Experimental Designs

Aside from the advantages described above, SMART designs also allow comparisons of different treatment options within the context of what happens in later stages. Other than a SMART design, a scientist has two alternatives for testing and building adaptive interventions: multiple, one-stage-at-a-time, randomized trials; or a randomized trial that compares fully-formed adaptive interventions.

Multiple, One-Stage-at-a-Time, Randomized Trials

Imagine that you have three primary research questions:

  • Which intervention should I provide initially?
  • Which intervention should I provide to responsive patients?
  • Which intervention should I provide to non-responsive patients?

By identifying the initial treatment first, and then separately identifying the best treatments for responders and non-responders, you could create an adaptive intervention, but you would lack some of the information provided in a SMART design.

Compared to multiple, one-stage-at-a-time, randomized trials, SMART designs provide

  • better ability to compare the impact of a sequence of treatments, rather than examining each piece individually,
  • increased ability to test tailoring variables, and
  • decreased impact of cohort effects (because non-responders in a SMART are more representative of all non-responders to initial treatment, including both highly motivated and less motivated individuals).

Randomized Trial of Fully-Formed Adaptive Interventions

By studying prior research, you may decide to start all patients on the least intensive treatment and increase treatment when necessary. This would also not provide you all the information that a SMART design could.

Compared to a randomized trial that compares fully-formed adaptive interventions, SMART provides the ability to understand what is working (and not working) within the intervention. SMART allows you to compare options including dose, type, and timing of treatments. Also, secondary analysis of data from a SMART allows researchers to assess tailoring variables.

For more information about the relative merits of SMART, see Lei et al. (2012).

Read an example of SMART intervention for relapsing alcoholics.

See our recommended reading for SMART.

Interested in other designs for behavioral interventions? Read about our work on optimizing behavioral interventions.

Reference

Lei, H., Nahum-Shani, I., Lynch, K., Oslin, D., & Murphy, S. A. (2012). A “SMART” design for building individualized treatment sequences. Annual Review of Clinical Psychology, 8, 14.1-14.28. doi: 10.1146/annurev-clinpsy-032511-143152 PMC Journal- In process

 

Last updated: May 11, 2020

Applet for calculating the minimum sample size for a pilot SMART

April 28, 2017:Applet for SMART

The Methodology Center is pleased to announce the availability of a new web applet for calculating the minimum sample size for a pilot SMART. The sequential, multiple assignment, randomized trial (SMART) is a novel experimental design that can be used to build high quality adaptive interventions that adapt to patient need. Pilot SMARTs can be used to examine feasibility and acceptability issues of adaptive interventions embedded in a full-scale SMART study.

Read more about SMART.

Open the applet.

Featured Article: Analyzing Data from a SMART to Prevent Alcohol Abuse Relapse

April 6, 2017:Ishani

Response to substance abuse treatment can look very different between individuals and even within individuals at different points in time. Sequential, multiple assignment, randomized trials (SMARTs) are being used to develop interventions that adapt based on individual needs and circumstances. New methods for data analysis show promise for improving intervention developers’ ability to tailor an intervention even more specifically to an in individual’s needs for a broad range of health issues, including substance use. In a recent article in the journal Addiction, Methodology Center researchers Inbal (Billie) Nahum-Shani, Daniel Almirall, and their collaborators demonstrate the utility of Q-learning, a method developed in computer science, for the analysis of data from a SMART to prevent relapse among individuals with alcohol use disorders. Q-learning helped the authors identify a subset of individuals who appeared to be responding to treatment, but who needed additional treatment to maintain progress.

The authors analyzed data from 250 participants in the Extending Treatment Effectiveness of Naltrexone (ExTEND) trial (D. Oslin, P.I.; NIAAA; R01 AA014851). ExTEND was a 24-week SMART that examined how to build an adaptive intervention to prevent relapse among people with alcohol use disorders using the drug naltrexone. Naltrexone is promising for treating alcohol dependence, but there is a broad range of responses to the drug. The researchers used data from the ExTEND SMART to construct an adaptive intervention that is tailored even further based on an individual’s response to the initial treatment.

The resulting adaptive intervention recommends additional treatment for a subset of participants who, though initially classified as responders to Naltrexone, are likely to benefit from a more intense maintenance intervention. Q-learning is similar to moderated regression analysis, but it is suitable for examining whether and how certain covariates are useful in developing or improving an adaptive intervention. Lead author Inbal “Billie” Nahum-Shani said, “Q-learning can help us identify new ways to tailor treatments beyond the tailoring variables we typically include in a SMART by design. The goal of this paper is to provide an accessible overview of this method to investigators in the area of substance use disorders and to demonstrate how it can help advance the science of adaptive interventions in this important field.”

By using Q-learning with data from a SMART, researchers can build empirically validated adaptive interventions for a broad array of health problems.

Open the article. (Journal access required.)

Reference

Nahum‐Shani, I., Ertefaie, A., Lu, X. L., Lynch, K. G., McKay, J. R., Oslin, D. W., & Almirall, D. (2017). A SMART data analysis method for constructing adaptive treatment strategies for substance use disorders. Addiction.

Grant: Expanding the Methodological Toolbox for Sequential, Multiple Assignment, Randomized Trials (SMARTs)

smartWORDLESSOctober 31, 2016:

Over the course of treatment, a clinician often alters treatment based on patient characteristics or response to earlier treatment. Sequential, multiple assignment, randomized trial (SMART) designs provide the data needed to construct high-quality adaptive interventions. Interventions that adapt at the right times (e.g., intensifying for people who do not respond to the initial treatment) can improve participant outcomes while decreasing the cost and burden of the intervention (e.g., stepping down treatment for responsive participants). SMART designs are currently being used around the world in dozens of trials to build adaptive interventions for drug use, HIV, ADHD, autism, obesity, and more.

Last year, a team of Methodology Center researchers was awarded a grant from the National Institute on Drug Abuse (R01 DA039901) to expand the methodological toolbox available for intervention designers seeking to analyze data and plan future SMART studies.

This research will develop multilevel models to allow intervention scientists to answer new questions using longitudinal data from a SMART—for example, to compare the effect of two adaptive interventions on changes in craving or substance use over time. The research will also develop sample size calculators to facilitate the planning of SMARTs studies with longitudinal outcomes. Co-principal investigators Inbal (Billie) Nahum Shani and Daniel Almirall are excited about the potential of this research to further expand the usefulness of SMART designs. Billie said, “Right now, the methods that we have for analyzing and planning sample size for SMART studies are relatively limited.  For example, they allow us to compare adaptive interventions only in terms of end-of-study outcome. However, many scientists are interested in taking advantage of longitudinal data they often collect in the course of a SMART study, and using it to compare adaptive interventions in terms of trajectories of change. For example, scientists may want to study change in HIV-risk behavior during the course of the intervention program. This is because modeling change, rather than end-of-study outcome, provides greater statistical power and a more nuanced picture of how the adaptive intervention works. In this project we will develop the tools to allow intervention designers conduct these analyses and plan future SMARTs with longitudinal outcomes.”

Other researchers on the team that include Linda Collins, John Dziak, and Susan Murphy. Billie, Danny, and Susan are based at University of Michigan, and Linda and John are at Penn State. This grant will add five more years of methodological research to the development of SMART, allowing this valuable method to be applied even more broadly.

Read more about SMART 

Year of SMART

September 7, 2016:

The Methodology Center is declaring this the “Year of SMART” to raise awareness among researchers, reviewers, and program officersbilliedanny at various agencies about the potential value of the sequential, multiple-assignment, randomized trial (SMART). SMART is a type of multi-stage factorial experimental design that allows researchers to build or optimize high-quality adaptive interventions. An adaptive intervention is a sequence of individually tailored intervention decision rules that help guide how best to adapt and re-adapt an intervention over time based on the evolving condition of the individual.

Scientists often have many questions about how best to develop the most effective adaptive intervention. This includes questions about how to initiate treatment (e.g., with more intensive treatment or with less intensive treatment), how best to monitor individuals’ progress in treatment (e.g., monitoring frequently), and what next treatment is best for individuals based on their response to the initial treatment. SMART has become a go-to approach to answer such critical questions. SMART extends a factorial experimental design to settings where there are multiple stages of treatment.

In recent years many funded research projects have been using SMART to build adaptive interventions for a broad range of health problems. For example, a recent special issue of the Journal of Clinical Child and Adolescent Psychology was devoted explicitly to the topic of adaptive interventions to improve the lives of children and adolescents with emotional or mental health disorders; various studies in this special issue use SMART.

Recent work on SMART by Center researchers focuses on developing new methods for multi-level adaptive interventions. Here, the goal is to understand whether or how best to intervene adaptively in classrooms, schools, clinics, or even organizations, with the goal to improve the health outcomes of the individuals within the clusters. For example, a current project is using SMART to understand how best to implement an evidence-based intervention for mood disorders in community-based mental health clinics. This implementation science study is a first-of-its-kind, NIMH-funded, cluster-randomized SMART. It randomizes clinics across Michigan and Colorado to different implementation interventions, with the goal of improving mental health outcomes for the individual patients within these clinics. A second IES-funded cluster-randomized SMART focuses on improving academic engagement outcomes of children with autism. In this SMART children are randomized to receive or not receive a classroom-level intervention, and then children not sufficiently responding are re-randomized to a peer-mediated or parent-mediated intervention.

Center researchers are also using SMART to develop adaptive mobile health (mHealth) interventions. mHealth tools offer many opportunities to improve the accessibility and scalability of care. The relatively low cost and widespread use of many mHealth tools make them ideal for use as minimal support in stepped-care interventions. A stepped-care intervention is a form of an adaptive intervention in which minimal (low cost, low burden) support is offered first, and then the extent of support (e.g., in terms of cost and burden) can be stepped up and down depending on response to initial support. However, many interesting questions arise concerning the most effective way to integrate mHealth tools in such stepped-care interventions. These include how best to initiate mHealth support (e.g., with or without additional human support); how to best assess non-response to mHealth support (e.g., how to identify early non-responders); and the best way to step care up and down based on response to mHealth support. An NIH-funded study is using SMART to investigate how to best initiate and augment mHealth tools for weight loss with more traditional intervention components (e.g., coaching, meal replacement) to treat obese adults.

We initiated this one-year effort to encourage each of you to consider novel methods for your research on adaptive interventions, including whether and how SMART could be used in your research. Stay tuned all year for the latest on SMART!

Read more about SMART and adaptive interventions.

New Grant: Smartphone-Based Interventions For Alcohol Abuse

November 23, 2015:
Congratulations to the team of researchers from The Methodology Center and North Carolina State University on their recently awarded grant fromliquorbottles the National Institute on Alcohol Abuse and Alcoholism, “Data-Based Methods for Just-in-Time Adaptive Interventions in Alcohol Use.” Despite the many problems associated with alcohol abuse, relatively few people in the United States receive treatment for alcohol use disorders.  Using smartphones to deliver interventions will allow more people to be treated, and will reduce the cost of treatment. This project aims to develop the methods and algorithms needed to provide each individual with the proper intervention exactly when he or she needs it.

The just-in-time adaptive intervention (JITAI) leverages the immense amount of data a smartphone or other mobile device can collect about a user’s activity level, location, and more. As these data are collected, JITAIs link these data in real time to the best treatment option, such as a behavioral intervention, a cognitive intervention, or social support. A JITAI responds to an individual’s specific circumstances to provide the needed treatment just in time. The JITAIs developed in this project will help prevent, treat, and promote recovery from alcohol use disorders.

Principal Investigator Susan Murphy states, “The focus of this grant is on data analytics to support real-time precision medicine. That is, these algorithms will allow us to personalize the JITAI dynamically as each individual goes about their life.   Two people may appear very similar in terms of their background and present situation.  But by tracking how each person responds to treatments, these analytics will allow us to personalize the JITAI to provide the right intervention at the right time and in the right setting for that person.”

Methodology Center Principal Investigator Susan Murphy and her team at the University of Michigan created and are developing JITAIs. The team for this project consists of experts in statistical methods (Susan Murphy, Eric Laber), computer science (Ambuj Tewari), and health behavior change and alcohol use (Inbal Nahum-Shani). The researchers will disseminate JITAIs and related innovations through peer-reviewed journal articles, an advanced workshop, and open-source software.

2014 Summer Institute: Experimental Design and Analysis Methods for Developing Adaptive Interventions: Getting SMART

Daniel AlmirallInbal Nahum-ShaniJune 19-20, 2014

Daniel Almirall and Inbal Nahum-Shani

 

 

Workshop objectiveSummer Institute 2014 participants

The overarching goals of this two-day workshop were to (a) provide an introduction to adaptive interventions; (b) help participants gain the background needed to plan a Sequential Multiple Assignment Randomized Trial (SMART); and (c) help participants learn how to implement data analytic methods with SMART study data to construct adaptive interventions.

General description

At this workshop, participants were provided with a hard copy of all lecture notes, select computer exercises and output, and suggested reading lists for future reference. Three different formats were used. First, all materials were presented following the standard didactic format with a slideshow. Second, there were practice exercises designed to help participants connect the material with their own research area.  These practice exercises were focused on SMART study design principles aimed at helping to prepare participants to write grant proposals that use a SMART design to build adaptive interventions. Third, there were computer exercises using SAS®. Computer code and simulated data examples were supplied by the instructors. The computer exercises helped investigators learn how to implement typical primary and secondary analyses using data arising from a SMART and to interpret the results. Throughout the workshop, time was set aside for Q&A and discussion about how the concepts learned in class can be applied in participants’ research. 

Topics that were covered:

  • When and why adaptive interventions are needed
  • How adaptive interventions differ from fixed (one-stage) interventions
  • The critical components of adaptive interventions: decision points, tailoring variables, intervention options and decision rules
  • The role of theory in developing adaptive interventions
  • Examples of adaptive interventions from the literature
  • The difference between a moderator variable and a tailoring variable
  • SMART study principles, including how to provide a rationale for designing a SMART
  • How SMARTs differ from standard randomized clinical trials
  • Different types of SMART designs
  • How to choose the sample size for a SMART (statistical power considerations)
  • Common types of primary and secondary scientific aims in a SMART
  • Data analytic strategies used to examine primary and secondary scientific aims in a SMART

In addition to the above topics, there were several hands-on computer exercises, open discussion times, and question/answer periods.

2014 Instructors

DANmirallDaniel Almirall, Ph.D. is a Research Assistant Professor at the Institute for Social Research at the University of Michigan who works with clinical scientists and health behavior researchers to design SMARTs. Dr. Almirall was a recipient of pilot funding from The Methodology Center and was mentored by Dr. Susan M. Murphy. In 2011-12 alone, Dr. Almirall gave over 25 presentations, workshops, and talks on SMART. His first-authored articles on methodology have appeared in top statistical journals such asStatistics in MedicineBiometrics, and Journal of the American Statistical Association; and in applied journals including Journal of Child and Adolescent Psychopharmacology and Prevention Science. His collaborative work has appeared in journals such as Journal of the American Medical Association and Journal of Studies on Alcohol and Drugs. Dr. Almirall was recently awarded an R03 from NIMH to develop a novel method for discovering tailoring variables in childhood mental health treatment research. For more information about Dr. Almirall, visit https://methodology.psu.edu/people/dalmirall.

 

IshaniInbal Nahum-Shani, Ph.D. is a Research Assistant Professor at the University of Michigan’s Survey Research Center within the Institute for Social Research. She works closely with Dr. Susan M. Murphy, creator of SMART, and has received pilot funding from the Methodology Center for her work on SMART. She currently has funding from multiple agencies, including the National Institute on Alcohol Abuse and Alcoholism (NIAAA), for her work on adaptive interventions. Dr. Nahum-Shani’s first-authored articles have appeared in outlets including Annual Review of Clinical PsychologyPsychological Methods, andMultivariate Behavioral Research. She has given over a dozen workshops and presentations on SMART in the last three years and continues to work on introducing the concept of adaptive interventions to behavioral and social scientists. For more information about Dr. Nahum-Shani, visit http://www-personal.umich.edu/~inbal/.

NIH Funding Announcement Calls for Sequential, Multiple Assignment, Randomized Trials (SMARTs)

January 17, 2013:
A new announcement from NIH seeks proposals that improve behavioral treatments for drug abuse, HIV, chronic pain, or related behaviors. PA-13-077 is sponsored by the National Institute on Drug Abuse (NIDA), the National Institute on Alcohol Abuse and Alcoholism (NIAAA), and the Office of Behavioral and Social Sciences Research (OBSSR). This program announcement specifically solicits proposals featuring sequential, multiple assignment, randomized trial (SMART) designs because of SMART’s applicability to efficacy studies and to translating interventions into real world settings.

SMART, an experimental design method for building adaptive health interventions, was developed by Methodology Center Principal Investigator Susan Murphy and her collaborators. Adaptive health interventions allow clinicians to create treatment sequences that change based on a patient’s characteristics or responses to earlier treatments. Adaptive interventions can improve outcomes for patients who are not responding to early treatments while decreasing burden and costs for patients who become stable during treatment. We are pleased by NIH’s recognition of how SMART designs can improve treatments and outcomes for patients.

View the program announcement

Read more about SMART