MTSS
Multi-Tiered Systems of Support (MTSS) and School Mental Health
Growing Adoption of MTSS:
There is an increasing need for rigorous evaluations of adaptive-sequential interventions within MTSS.
Few investigations have empirically examined the continuum of supports within and across tiers.
Prevention approaches with distinct theoretical foundations (e.g., PBIS, SEL) are available within and across tiers.
School-based practitioners need evaluations regarding optimal treatment sequencing.
Adaptive Treatment Strategies (ATS):
ATS are a natural fit within the MTSS framework.
Sequential multiple assignment randomized trials (SMART) offer a promising empirical approach to develop and compare adaptive treatment regimens.
Introduction to School Mental Health
Increasing Focus: In recent years, there has been an increasing focus on school mental health and addressing "non-academic barriers to learning."
Prevalence of Mental Health Services: Approximately 20% of students receive some form of school mental health service, with continued growth in recent years (Foster et al., 2005).
Under-Identification of Mental Health Challenges: Mental health challenges remain frequently under-identified.
This makes systems-level, school-wide mental health promotion and prevention efforts critical (Flett & Hewitt, 2013).
Diversity of Student Needs: Student needs are diverse, ranging from internalizing problems and substance use problems to externalizing problems.
Limitations of Traditional Referral Systems:
Referrals to community healthcare agencies are time-consuming and expensive.
They do not readily translate into interventions or accommodations that can be offered in school settings.
Limitations of Mandated School Services:
Special education and alternative learning placements require special qualifications and are costly.
They are available only to students with the most serious behavioral and emotional problems.
Schools Taking Ownership: Schools have increasingly adopted multi-tiered systems of support (MTSS) to provide proactive, comprehensive, and evidence-based supports.
MTSS Framework:
Typically conceptualized as a three-tiered model.
Provides layered interventions that begin with universal, school-wide programming.
Interventions increase in intensity and differentiation depending on the students' response to preceding interventions (Fletcher & Vaughn, 2009).
Examples of MTSS Models: Response to Intervention (RTI) and Positive Behavior Interventions and Supports (PBIS).
These models apply a systematic and empirically-driven MTSS framework to ensure that students receive more timely and effective services (Fuchs & Fuchs, 2006; Hawken, Vincent, & Schumann, 2008).
MTSS Applied to Student Behavior
Tier 1 Interventions:
Generally consist of a school-wide code of behavioral expectations that are explicitly taught to all students and reinforced.
All students, regardless of their degree of risk, are exposed to a general classroom management system including clear behavioral expectations and supports (i.e., universal intervention).
Tier 2 Interventions:
Students showing an inadequate response (i.e., continue to display behavioral problems) are stepped-up to targeted and more intensive Tier 2 interventions.
Typically consist of more focused support programs that are often delivered in a small group format.
Examples include manualized programs like Coping Power (Lochman & Wells, 2002), social skills training, or efficient individual interventions such as behavior contracts or Check-in/Check-out (CICO).
Tier 3 Interventions:
Students who are unresponsive to small group intervention and continue to struggle with their behavior are stepped-up to Tier 3 interventions.
These are the most intensive and often provide function-based individualized behavioral intervention plans or involve referral for special education services (Crone, Horner, & Hawken, 2004).
Additive Nature of Tiered Supports: Lower-level supports are still available to students requiring support at higher tiers.
Critical Components of MTSS:
Monitoring of students' response to the interventions with data-based measures.
Establishing criteria for transitioning between levels of support (Gresham, 2005; Sugai, Horner, & Gresham, 2002).
Opportunity to Support Student Mental Health: As the implementation of MTSS continues to proliferate in educational settings, there exists significant opportunity to support student mental health in ways not previously realized.
Advocacy and Federal Directives: Advocacy and federal directives for providing students with school-based mental health services have reinforced this movement in addressing the mental health needs of students (U.S. Department of Education, 2003).
Delivery of Evidence-Based Programs: A foundational component of the MTSS framework involves the delivery of evidence-based programs.
Increasing Pressure on Schools: There has been increasing pressure placed on schools to import evidence-based prevention and treatment programs in response to students' mental health needs (Langley, Nadeem, Kataoka, Stein, & Jaycox, 2010).
Growing Number of Evidence-Based Programs: There are a growing number of evidence-based programs established for use in school settings (Forman et al., 2013).
Focus of Programs: These programs typically address behavioral, social, and emotional factors assumed to cause or exacerbate disruptive, noncompliant and aggressive behavior (Wilson & Lipsey, 2007), although a growing number address mental health more broadly (e.g., emotion regulation, trauma, depression, anxiety).
Modalities of Programs: Feature a variety of modalities including classroom-wide support systems and behavioral health curricula, small group socio-emotional skills training and peer support, and comprehensive, multicomponent programs that typically integrate training for child, parent, and teacher (August, Bloomquist, Realmuto, & Hektner, 2007; August, Realmuto, Winters, & Hektner, 2001).
Standardized Delivery: To standardize and facilitate delivery, these programs are generally delivered with uniform composition, dosage, and duration to students regardless of their individual risks and needs (August, Gewirtz, & Realmuto, 2010).
Limitations of "One Size Fits All" Approach: This approach assumes that all children have similar needs.
Modest Effect Sizes: Despite their intuitive appeal and evidence base, such programs have yielded only modest effect sizes with considerable variability in individual response (Rones & Hoagwood, 2000).
Call for More Adaptive Approaches: Such performance has led some researchers to call for more adaptive, customized approaches that are tailored to the individual needs of youth (Collins, Murphy, & Bierman, 2004).
Tailored Problem-Solving Approach: Adopting a more tailored problem-solving approach to service delivery is consistent with the basic tenets of MTSS as a proactive and responsive framework, yet efficiency and feasibility are also very real and important concerns.
Avoiding "Program for Every Problem" Phenomenon: We must also avoid the “program for every problem” phenomenon (Domitrovich et al., 2010).
Challenges in Delivering Tailored Approach: Determining how to deliver a tailored, problem-solving approach while maintaining efficiency and feasibility is a challenge.
Guidelines Needed: While this framework offers a promising approach for providing students with the services they need, there are few guidelines for:
(a) selecting the most appropriate interventions for each tier
(b) determining how best to sequence the interventions in a tiered approach
(c) how to determine the best intervention sequence for any individual student.
Differing Theoretical Orientations: These challenges are compounded by the emergence of programs developed with differing theoretical orientations.
For example, interventions implemented within the context of PBIS are grounded in behavioral principles, while interventions implemented within the context of social-emotional learning (SEL) are grounded in the principles of positive youth development.
Difficulty for School Professionals: These challenges make it incredibly difficult for school professionals to determine which programs to implement in their settings, and which programs will yield the greatest effects for their student population.
Emerging Innovation: Adaptive Treatment Strategies (ATS): The present article describes an emerging innovation in the development and validation of precision-based interventions for youth who experience social, emotional, and behavioral impairments and need additional support.
ATS (also known as dynamic treatment regimes) apply principles similar to those used in MTSS to tailor each individual's intervention over time based on assessment of ongoing response but extend these models in several ways.
ATS Specifics:
ATS specify (a) which intervention option to offer first
(b) at what time point response should be assessed and interventions adjusted
(c) which intervention option should be offered if there is nonresponse to the first intervention option.
Intervention options may vary in intensities, types, and/or modalities.
Sequential Multiple Assignment Randomized Trials (SMART): The construction of these decision rules is aided by an innovative research methodology called sequential multiple assignment randomized trials (SMART).
SMART empirically evaluates multiple intervention sequences and associated decision rules within a single trial in order to identify optimal ATS.
Article Overview: In the text that follows we:
(a) present the rationale for ATS
(b) describe the SMART technology used to operationalize ATS
(c) describe a SMART prototype currently being delivered by a community agency to preempt the development of conduct disorder among at risk youth
(d) describe an example of how schools might apply a SMART to evaluate multi-tiered interventions to prevent or deescalate behavior problems.
Adaptive Treatment Strategies (ATS)
ATS Defined: ATS use ongoing information about an individual (e.g., changes in behavioral status) to make subsequent intervention decisions through the use of decision rules (i.e., algorithms).
Decision Rules: Decision rules specify how the composition and/or intensity of an intervention should be adjusted at critical decision points such as when an individual is not responding to a current intervention (August, Piehler, & Bloomquist, 2014; Lei et al., 2012).
Resemblance to Real-World Practice: ATS resemble ‘real-world’ practice where practitioners often change treatments when an individual fails to demonstrate a desired response, absent empirically established decision rules.
Recommendations Based on Individual Characteristics: With ATS, recommendations for adjusting treatment are based on individual characteristics that are collected and measured during treatment, such as “has the individual exhibited significant symptom reduction” or “has the individual reached a specified level of adaptive functioning?”
Readjusting Intervention Plan: When an individual displays no response or possibly a suboptimal response, practitioners may readjust the intervention plan by increasing dosage or switching to a different intervention.
Time-Varying Approach: This time-varying approach is particularly useful for the treatment of chronic disorders such as depression (Lavori, Dawson, & Rush, 2000), Attention-Deficit Hyperactivity Disorder (ADHD; Pelham et al., 2016), and alcohol and drug dependence (Krantzler & McKay, 2012).
Prevention Tool: ATS may also have a role as a prevention tool by redirecting risk trajectories for children or youth deemed to be at heightened risk for conduct disorder (August et al., 2014).
Natural Fit with MTSS: Systematically incorporating empirically-derived ATS within the MTSS framework appears to be a natural fit.
With both approaches, the goal is to adapt intervention options based on an individual's progress to a specific goal.
Four-Stage Problem-Solving Model: Underlying the MTSS framework, an individually-oriented four-stage problem-solving model is frequently utilized (Kratochwill & Bergan, 1990).
These stages entail identification of a problem, problem analysis, intervention implementation, and intervention evaluation.
Careful Selection of Intervention Programming: This problem-solving process allows for careful selection of appropriate intervention programming based on student needs and evaluation of program effects.
ATS as a Systematic Extension: Indeed, in many ways ATS could be considered a systematic extension of this approach within MTSS.
Value Added to Incorporating ATS Paradigm: We propose that there is value added to incorporating the ATS paradigm as a mechanism for studying MTSS.
Consistency with MTSS Rationale: ATS is consistent with the underlying rationale for MTSS, with a central aim being to ensure that all youth are able to access the services they need.
Evaluating Interventions in Isolation: In particular, within the context of MTSS, researchers often evaluate evidence-based interventions in isolation as standalone programs, with little attention to how a series of interventions perform either within tiers or in combination across tiers of support (i.e., synergistic effects).
Disconnect Between Research and Delivery: Thus, a disconnect exists between how these interventions are evaluated and how they are delivered within MTSS.
Need to Bridge Research-to-Practice Gap: Such research is imperative in order to help bridge the research-to-practice gap and is consistent with growing interest in implementation science in school settings (Owens et al., 2014).
Theoretically-Informed Intervention Planning: Furthermore, the MTSS emphasis on theoretically-informed intervention planning is highly germane to the development and implementation of ATS.
Problem Analysis Stage: Within MTSS, a problem-solving approach is often implemented, including a problem analysis stage, which involves the identification of specific, potentially malleable, causal factors maintaining a problem (Kratochwill & Bergan, 1990).
Informing Intervention Selection: In turn, this analysis informs the selection of more appropriate and precise interventions.
ATS and Intervention Programs: Similar to intervention selection with the problem-solving approach, intervention programs utilized within ATS typically consist of interventions that target common theoretically-derived mechanisms of problems.
Empirically-Derived Decision Rules: Within ATS however, empirically-derived decision rules utilizing individual characteristics may guide the selection of intervention programming.
Example Assessment of Problem Behavior: For example, an assessment of problem behavior may reveal specific established etiological pathways (e.g., social skill deficits or poor emotion regulation) contributing to such behaviors in an individual student.
Decision Rule Guided Selection: A decision rule will guide the selection of an ATS that employs an intervention empirically demonstrated to most effectively target the identified key contributing factors.
Theoretically-Derived Decision Making: Thus, ATS allow for a process of theoretically-derived decision making that is guided by empirical evidence.
Student Response to Intervention: Finally, a hallmark of MTSS and the associated problem-solving approach involves student response to intervention, yet very little is known regarding non-responders in the current landscape.
Research on Non-Responders: Research examining more precise and appropriate interventions for students who are unresponsive to intervention is critical to advance prevention efforts.
Pioneering Work in Reading: A number of researchers have pioneered work in this vein, particularly in the area of reading (e.g., Connor et al., 2009; Torgesen, 2000).
Less Attention to Social, Emotional, and Behavioral Domains: However, in educational settings, social, emotional and behavioral domains have received considerably less attention.
Evidence-Based Practices for Non-Responsive Students: It could be argued that, in many ways, evidence-based practices do not currently exist for non-responsive students.
Little Attention to Non-Responders in Traditional Research: Indeed, within traditional group-based research methodology, little attention is typically paid to non-responders.
Positive Main Effect Concerns: That is, when an intervention is found to produce a positive main effect, there are likely students participating in the intervention group who either (a) did not change or (b) actually declined in performance.
Concern for These Students: We should attend to and be concerned about these students and seek to develop more adaptive, precise and effective interventions.
Precision-Based Care: Such precision-based care is at the conceptual core of an MTSS framework: many consider MTSS to be a needs-driven, equity based approach to ensuring that all children receive the supports they need to be successful.
Trial-and-Error Approach: Instead, in practice, a trial-and-error approach to modifying subsequent efforts is typically employed.
Elements School-Based Teams Change: For example, school-based teams my decide to change any number of elements, including:
(a) the format of delivery for students unresponsive to intervention (small group versus individual)
(b) dosage frequency or intensity
(c) implement a different intervention entirely.
Promising Approach: Adaptive treatment strategies offer a promising and rigorous approach to the study of intervention delivery and tailoring to promote positive outcomes for all students, but particularly for those with the greatest need.
SMART Technology
ATS Implementation in Schools: While schools implementing MTSS might employ ATS in their delivery, the embedded ATS may not have undergone rigorous evaluation and optimization.
Questions to Address in Constructing ATS: In constructing ATS, questions that need to be addressed include the best sequencing of interventions when individuals are not responding and the best time to evaluate response.
SMART Design: Construction of such high-quality ATS can be achieved with an innovative type of research design referred to as sequential multiple assignment randomized trials (SMART; Almirall et al., 2014; Lavori & Dawson, 2008; Murphy, Oslin, Rush, & Zhu, 2007).
Multiple Stages and Randomizations: A SMART design is implemented in multiple stages with individuals randomized multiple times to various intervention options across stages (see Lei et al., 2012).
Balanced Groups: Sequenced randomizations ensure that at each decision point, the groups of participants assigned to each of the intervention options are balanced in terms of participant characteristics.
Randomization Example: In the examples provided below each youth is randomized twice, initially and then again once it is known whether the youth is a responder or a non-responder to the initial intervention.
SMART Experimental Design & Randomization in ATS: It is important to note that even though the SMART experimental design involves randomization, once ATS have been developed, their delivery in customary practice does not involve randomization.
Key Time Points: Each randomization stage within the SMART becomes in the ATS a key time point where a decision is made as to whether to adjust the intervention.
Empirically-Derived Decision Rules: Data from the SMART are used to construct empirically-derived decision rules that may be utilized within these time points as a part of the ATS.
Selection of Appropriate Interventions: The selection of appropriate interventions is critical in the creation of effective ATS.
Possible Intervention Options: Possible intervention options may reflect different types of conceptual orientations (behavioral-based contingency systems vs. socio-emotional skills training), different intervention foci (youth vs parent), different modes of delivery (classroom, small group, individual, parent/family), different levels of dosage/intensity, and/or different approaches to increase engagement and adherence to the intervention.
Operationalizing Decision Rules: The goal is to operationalize decision rules such as—begin with intervention type X, if the individual shows favorable response, step-down to maintenance; if the individual shows a poor response, step-up to intervention Y.
Multiple ATS Derived from One SMART: Depending on the number of decision points and the number of intervention options to consider at each decision point, multiple ATS can be derived from any one SMART.
Answering Key Tactical Questions: Set up in this way, a SMART allows researchers to answer key tactical questions, such as “What is the best first stage intervention option,” “What second stage intervention option is best for individuals who do not show satisfactory response to the first stage intervention option,” “Which sequence of intervention options yields the best outcomes?”
Examination of "Downstream" or Synergistic Effects: This approach allows for an examination of “downstream” or synergistic effects whereby an initial intervention component enables an individual to benefit more substantially from subsequent intervention components (Murphy et al., 2007).
Systematically Identifying Effective Approaches: By evaluating different first-stage, second-stage, and overall sequences of interventions, a SMART allows us to systematically identify which approach may be most effective in addressing the diverse needs and risk factors of a population.
Tailoring Variables: In addition to tailoring intervention sequences based on assessment of ongoing response, SMART can further increase precision by identifying potential tailoring variables that reflect pre-intervention individual characteristics.
Measuring Personal Characteristics: By measuring personal characteristics before initiating services, researchers may evaluate these characteristics for their utility in predicting differential response to the various intervention sequences.
Secondary Tailoring Variables: If successfully identified, these personal characteristics could then serve as secondary tailoring variables to match individuals with their optimal intervention.
Potential Secondary Tailoring Variables: Potential secondary tailoring variables may include gender, age, SES, as well as biomarkers, personality traits and psychosocial risk factors.
Integrating Secondary Tailoring Variables: When validated in SMART, these secondary tailoring variables can be integrated into the derived ATS.
Answering the Question: “What Works Best for Whom?”: As such, these data help answer the question, “What works best for whom?”
Application of ATS Within and Across MTSS Tiers: ATS derived through SMART may be applied both within and across MTSS tiers.
Within Tiers: Within tiers, an ATS may inform the section of an optimal intervention strategy for a particular youth at that level of support. Strategies within tiers may be selected based on knowledge of a best intervention option within a combination of interventions offered. Individual tailoring variables may also be used to select an optimal intervention strategy for a particular youth within a tier of support.
Across Tiers: Across tiers, ATS will provide a standardized approach to evaluate non-response within a particular tier and the appropriate subsequent more intensive intervention to be offered at a higher tier.
Consistent Empirical Guidance: By providing decisions rules dictating intervention selection, identification of non-responders, and intervention sequence, ATS have the potential to build more consistent empirical guidance across the MTSS framework.
Executing a SMART Design
Juvenile Diversion Agency: The Community Prototype
Description: This section presents a description of a community-based implementation of a SMART design (see August et al., 2014).
Intent: The intent is to provide the reader with the rationale and approach for constructing ATS as a programming framework for a high risk youth population (diversion youth) and an illustration of how a SMART design was crafted to operationalize ATS.
Collaborative Partnership: A collaborative partnership between a youth-serving agency and a university-based team of research investigators implemented this project.
Agency Role: The agency serves the county attorney's office by providing pre-court diversion programming for juvenile offenders who have been cited by law enforcement for various status and misdemeanor offenses including shoplifting, vandalism, disorderly conduct, underage drug use, and assault but have not yet established a pattern of serious and chronic antisocial behavior or have been formally adjudicated.
Diversion as a Portal: Diversion serves as a key portal for identifying at-risk youth, many of whom are at heightened risk for developing Conduct Disorder (CD) as well as depression, anxiety disorders, and substance use disorders (Wareham, Dembo, Poythress, Childs, & Schmeidler, 2009).
Standard Practices for Diversion Agencies: The standard practices for diversion agencies are to require restitution, community service, or simply to warn-and-release.
Limitations of Punitive Interventions: Although these punitive interventions satisfy the public demand for accountability, there is little evidence that they prevent escalation of conduct problems or progression to more serious mental disorders (Patrick & Marsh, 2005; Wilson & Hoge, 2013).
Alternative: Life-Skills Approach: An alternative to restorative justice programs is a life-skills approach that emphasizes the acquisition of strengths and the building of human capital in order to maximize the likelihood that youth offenders will veer away from a delinquent lifestyle toward more conventional goals (Guerra, Williams, Tolan, & Modeski, 2008).
Emphasis on Evidence-Based Programs: In light of the increasing emphasis on evidence-based programs for prevention and treatment of youth conduct problems, the collaborative community agency/research team opted to select interventions with an evidence-base for reducing conduct problems.
Effective Interventions: There is overwhelming evidence that youth problem-solving skills training and/or parent behavioral management skills training are effective interventions in this domain (Kazdin, 2010).
Challenges Faced by Diversion Counselors: The collaborative further recognized that diversion counselors confront additional intervention challenges as a result of the heterogeneity in the diversion population.
Heterogeneity in Diversion Population: The diversion population includes youth who may be one-time offenders and at minimal risk for future offending, periodic offenders who are at moderate risk for future offending, as well as chronic offenders who are at heightened risk for serious and chronic offending.
Limitations of "One Size Fits All" Approach: Consequently, conventional interventions that apply a “one size fits all approach” are unlikely to be effective in addressing these diverse individuals.
Negative Effects of Incongruent Interventions: Moreover, providing interventions to youth that are incongruent with their needs and motivations can result in low levels of engagement and dropout (Kazdin & Wassell, 1999) as well as negative peer contagion effects particularly when programs are provided using a group format (Dodge, Dishion, & Lansford, 2006).
Need for New and Creative Approaches: It stands to reason that new and creative approaches are needed.
Guidelines for Effective and Efficient Intervention Framework: With these considerations in mind, the collaborative produced the following guidelines to assist in their efforts to produce an effective and efficient intervention framework:
Youth (and their families) referred by law enforcement for juvenile diversion programming are not help-seekers in the traditional sense. Thus, youth and their families must perceive intervention options offered to them as relevant to their needs and administered with minimal burden.
In light of the varying degrees of risk for continued offending among a diversion population (heterogeneity), an adaptive intervention approach in which intervention options are tailored to the youths' risk profiles may produce the best outcomes.
An efficient and cost-effective adaptive intervention approach would feature a sequential, stepped care delivery system in which youth receive only the intervention(s) they need.
In order to construct optimal adaptive treatment strategies (ATS), SMART design technology will be used. Youth are randomized to two brief-type intervention options at the first tier. Responders are stepped-down and monitored over time while non-re- sponders are stepped-up and randomized to more intensive interventions at the second tier.
In lieu of conventional punitive interventions, diversion counselors will offer strength-based skills training interventions with the primary orientation being motivational enhancement coupled with the teaching of decision-making skills.
Because youth-focused and parent-focused skills training modalities have been validated in previous prevention and treatment research addressing conduct problems, both foci will be compared across intervention tiers.
In order to assess intervention response, diversion counselors will use an empirically-based response assessment tool. The tool will include measures that assess the youth's risk trajectory leading to conduct disorder.
Interventions
Selection: The collaborative selected two evidence-based intervention options to construct the ATS.
Teen Intervene (TI): The Teen Intervene program (TI; Winters & Leitten, 2007) is a youth-focused intervention that includes motivational enhancement, prosocial goal-setting, and training in responsible decision-making and social problem-solving with the goal of choosing attitudes and behaviors that are healthier alternatives to antisocial behaviors.
Everyday Parenting (EP): The Everyday Parenting program (EP; Dishion, Stormshak, & Kavanagh, 2011) is a parent/family-focused intervention that addresses three broad areas of parent/family skills building:
(a) behavioral management support in the form of contingent positive reinforcement and punitive consequences to help regulate adolescent behaviors
(b) limit setting and supervision of youths' activities, whereabouts, and peer affiliations to minimize opportunities for inappropriate or dangerous risk-taking
(c) family interaction skills to facilitate parent-adolescent communication and problem-solving to facilitate responsible decision-making.
Brief and Extended Formats: Both programs can be modified to be delivered in “brief” and “extended” formats.
The brief versions include two or three sessions and include motivational-interviewing concepts that may be especially effective as first-tier intervention options, particular for youth at low risk for escalation of conduct problems (Jensen et al., 2011).
Extended models are best suited for youth at moderate to higher degrees of risk, include an additional three-to five sessions, and provide intensive skills training.
Response Assessment Tool
ATS Tailoring: As noted above, ATS tailor treatment via decision rules that specify how the intensity or type of intervention should be adjusted depending on individual characteristics that indicate a satisfactory response to the current intervention.
Measuring Responsiveness: Measuring students' responsiveness and establishing criteria for transitioning between tiers is critical to the performance of a SMART design.
Treatment-Based Intervention Systems: For treatment-based intervention systems delivered in clinic setting, the logical candidate would be a reduction in problem behaviors (e.g., non-compliance, fighting) or functional impairments (e.g., poor peer interactions, academic difficulties).
Prevention Systems: For prevention systems that target high risk, but asymptomatic individuals, the response indicators are not readily apparent. Successful response may reflect a change in a youth's risk trajectory (e.g., motivation to change behavior, attitudes toward aggression, norms about the appropriateness of aggressive behavior).
Multi-Dimensional Risk Assessment Tool: To determine intervention responder status with the diversion sample, we used a multi-dimensional risk assessment tool that included measures that reported:
(a) current conduct problems
(b) current functional impairment
(c) deviant peer affiliations.
Criteria for Additional Services (Non-Responder): To be flagged as at-risk and in need of additional services (i.e., non-responder) following completion of Tier 1 intervention, youth would need to show evidence of any of one of the following criteria:
(a) problematic conduct problems > 1 standard deviation (T-Score > 60) as rated by parents on the Behavioral Assessment System for Children ([BASC-2]; Reynolds & Kamphaus, 2004)
(b) impaired adaptive functioning as rated by counselors on the Child and Adolescent Functional Assessment Scale (CAFAS [T-Score > 60]; Hodges, 2000)
(c) elevated exposure to deviant peer influences on the Friendship Scale (T-Score > 60; Child and Family Center, 2013a, 2013b).
Remaining at Elevated Risk: It is important to keep in mind in this example that youth may display some improvement (improve from three to one criterion) but nevertheless remain at elevated risk for serious conduct problems.
SMART Design Analog: Below we present a visual analogue of a SMART design that was employed in this study. At Stage 1, youth are randomized to either TI-Brief or EP-Brief. Responders to either of the Stage 1 options are stepped-down and monitored over time for maintenance of intervention effects. Non-responders to either of the Stage 1 options are stepped up and re-randomized to tier 2 options, either (a) continuation of the Stage 1 option with increased dosage (TI-Extended or EP-Extended), or (b) switched to the alternative extended intervention option.
Embedded ATS: Based on the number of stages and the number of intervention options, one or more ATS can be embedded within a SMART. The present SMART yields the following four ATS:
Youth-Only Skills Training ATS: Begin with TI-Brief, youth exhibiting a positive response to initial TI-Brief are stepped-down to monitoring, youth exhibiting nonresponse are stepped up to TI-Extended (this is a youth-continuation ATS).
Youth Skills Training then Parent Support ATS: Begin with TI-Brief, youth exhibiting a positive response to initial TI-Brief are stepped-down to monitoring, youth exhibiting nonresponse are stepped-up to EP-Extended (this is a begin with youth then switch to parent ATS).
Parent-Only Support ATS: Begin with EP-Brief, youth exhibiting a positive response to initial EP-Brief are stepped-down to monitoring, youth exhibiting nonresponse are stepped-up to EP-Extended (this is a parent-continuation ATS).
Parent Support then Youth Skills ATS: begin with EP-Brief, youth exhibiting a positive response to initial EP-Brief are stepped- down to monitoring, youth exhibiting nonresponse are stepped-up to TI-Extended (this is a parent then switch to youth ATS).
Key Questions Addressed: This SMART design permits several key questions to be addressed:
Which Stage one intervention provides the best response and thus should be offered initially?
Which Stage two intervention provides the best second tier intervention for youth who show non-response to a Stage one intervention?
Which sequential approach provides the best overall response – one that begins with a youth-focused intervention or one that begins with a parent/family-focused intervention?
Relevance to School Settings: While this SMART was implemented in a community context with diversion-referred youth, several aspects of this population and design are highly relevant to school settings.
Heterogeneity in School Settings: First, the significant heterogeneity of the population is also reflected in most school settings. Like diversion youth, students in the classroom may display similar behaviors but have wide range of underlying risk factors maintaining those behaviors. For this reason, incorporating a range of intervention intensities and foci in an ATS is essential to meet these diverse needs.
Challenges Engaging Parents: Second, effective engagement of parents in interventions often presents substantial challenges. Parents of di- version youth may resist participating in an intervention because of their perception that the child's offense was “not the parents' fault.” Parents of youth exhibiting behavioral problems in a school setting may be similarly resistant to involvement due to their perception of a lack of responsibility for their child's behavior when at school.
Motivational Enhancement and Minimizing Burden: Utilizing motivational enhancement and minimizing parental burden of service delivery are two beneficial strategies that can be incorporated into an ATS to increase parental engagement and reduce resistance to services.
Elementary School Context: The School Prototype
Problem-Solving MTSS Models: As noted above, problem-solving MTSS models such as RTI and PBIS are currently in vogue as a replacement for the traditional “Refer-Test-and Place” model commonly applied by schools to assist high risk youth (Cash & Nealis, 2004).
Refinement via SMART: Refinement of these problem-solving models can be informed by innovative research methodologies such as SMART that are currently being employed to construct ATS for individuals suffering with chronic and severe mental health and substance use disorders (Murphy, Lynch, Oslin, McKay, & TenHave, 2007; Shortreed & Moodie, 2012).
Limited Examples in School Settings: As this research with SMART is nascent, there are relatively few examples to guide its application in school settings.
Existing SMART Designs: Existing SMART designs employed in schools have typically focused on children with existing mental health diagnoses rather than more general at-risk populations.
Pelham et al. (2016) SMART Study: While some studies in this area are currently underway, a recently completed SMART study by Pelham et al. (2016) provides an example of a partially school-based implementation. This SMART evaluated options for sequencing and dosage of medication and behaviorally-based interventions for elementary school children with ADHD. The study incorporated school-based assessments of intervention response as well as teacher consultation regarding classroom based behavioral management practices.
Feasibility of Teacher-Reported Data: The study provides evidence for the feasibility of utilizing teacher- reported data and observed classroom behavior in determining response within a SMART as well as deploying aspects of SMART-embedded ATS within school settings.
Heuristic Framework: In this section we present a heuristic framework that illustrates how a SMART design might be integrated
Multi-Tiered Systems of Support (MTSS) and School Mental Health
Growing Adoption of MTSS:
There is an increasing need for rigorous, comprehensive, and longitudinal evaluations of adaptive-sequential interventions within MTSS to ensure their effectiveness and sustainability.
Few investigations have empirically examined the continuum of supports within and across tiers, hindering the understanding of how different interventions interact and impact student outcomes.
Prevention approaches with distinct theoretical foundations (e.g., PBIS, SEL) are available within and across tiers, creating a need for comparative studies to determine which approaches are most effective for specific student populations and contexts.
School-based practitioners need evaluations regarding optimal treatment sequencing to make informed decisions about which interventions to implement and in what order to maximize positive outcomes for students.
Adaptive Treatment Strategies (ATS):
ATS are a natural fit within the MTSS framework, providing a flexible and data-driven approach to tailoring interventions to individual student needs.
Sequential multiple assignment randomized trials (SMART) offer a promising empirical approach to develop and compare adaptive treatment regimens, allowing researchers to identify the most effective intervention sequences for different student profiles.
Introduction to School Mental Health
Increasing Focus: In recent years, there has been an increasing focus on school mental health and addressing "non-academic barriers to learning," recognizing the critical role of mental health in student success.
Prevalence of Mental Health Services: Approximately 20% of students receive some form of school mental health service, with continued growth in recent years (Foster et al., 2005), reflecting the increasing awareness and prioritization of mental health support in schools.
Under-Identification of Mental Health Challenges: Mental health challenges remain frequently under-identified, leading to delayed intervention and potentially exacerbating the impact on students' academic and social-emotional well-being.
This makes systems-level, school-wide mental health promotion and prevention efforts critical (Flett & Hewitt, 2013) to create a supportive and proactive environment for all students.
Diversity of Student Needs: Student needs are diverse, ranging from internalizing problems (e.g., anxiety, depression) and substance use problems to externalizing problems (e.g., aggression, defiance), requiring a comprehensive and tailored approach to mental health support.
Limitations of Traditional Referral Systems:
Referrals to community healthcare agencies are time-consuming and expensive, creating barriers to access for many students and families.
They do not readily translate into interventions or accommodations that can be offered in school settings, limiting the ability to provide immediate and integrated support for students' mental health needs.
Limitations of Mandated School Services:
Special education and alternative learning placements require special qualifications and are costly, restricting their availability and potentially creating disparities in access to mental health services.
They are available only to students with the most serious behavioral and emotional problems, leaving many students with subthreshold mental health needs without adequate support.
Schools Taking Ownership: Schools have increasingly adopted multi-tiered systems of support (MTSS) to provide proactive, comprehensive, and evidence-based supports for students' academic, behavioral, and social-emotional needs, including mental health.
MTSS Framework:
Typically conceptualized as a three-tiered model, with each tier representing a different level of intervention intensity and support.
Provides layered interventions that begin with universal, school-wide programming designed to promote positive mental health and prevent mental health problems for all students.
Interventions increase in intensity and differentiation depending on the students' response to preceding interventions (Fletcher & Vaughn, 2009), ensuring that students receive the appropriate level of support based on their individual needs.
Examples of MTSS Models: Response to Intervention (RTI) and Positive Behavior Interventions and Supports (PBIS), which provide structured frameworks for implementing MTSS in schools.
These models apply a systematic and empirically-driven MTSS framework to ensure that students receive more timely and effective services (Fuchs & Fuchs, 2006; Hawken, Vincent, & Schumann, 2008), promoting data-based decision-making and continuous improvement in service delivery.
MTSS Applied to Student Behavior
Tier 1 Interventions:
Generally consist of a school-wide code of behavioral expectations that are explicitly taught to all students and consistently reinforced to create a positive and predictable school climate.
All students, regardless of their degree of risk, are exposed to a general classroom management system including clear behavioral expectations and supports (i.e., universal intervention) to promote positive behavior and prevent behavioral problems.
Tier 2 Interventions:
Students showing an inadequate response (i.e., continue to display behavioral problems) are stepped-up to targeted and more intensive Tier 2 interventions designed to address specific behavioral needs.
Typically consist of more focused support programs that are often delivered in a small group format to provide individualized attention and skill-building opportunities.
Examples include manualized programs like Coping Power (Lochman & Wells, 2002), social skills training, or efficient individual interventions such as behavior contracts or Check-in/Check-out (CICO) to address a range of behavioral challenges.
Tier 3 Interventions:
Students who are unresponsive to small group intervention and continue to struggle with their behavior are stepped-up to Tier 3 interventions, the most intensive level of support.
These are the most intensive and often provide function-based individualized behavioral intervention plans or involve referral for special education services (Crone, Horner, & Hawken, 2004) for students with significant behavioral and emotional needs.
Additive Nature of Tiered Supports: Lower-level supports are still available to students requiring support at higher tiers, ensuring that students receive a comprehensive and integrated system of support.
Critical Components of MTSS:
Monitoring of students' response to the interventions with data-based measures to track progress and make informed decisions about intervention adjustments.
Establishing criteria for transitioning between levels of support (Gresham, 2005; Sugai, Horner, & Gresham, 2002) to ensure that students receive the appropriate level of intervention intensity based on their individual needs and progress.
Opportunity to Support Student Mental Health: As the implementation of MTSS continues to proliferate in educational settings, there exists significant opportunity to support student mental health in ways not previously realized, by integrating mental health promotion, prevention, and intervention strategies into the MTSS framework.
Advocacy and Federal Directives: Advocacy and federal directives for providing students with school-based mental health services have reinforced this movement in addressing the mental health needs of students (U.S. Department of Education, 2003), highlighting the importance of prioritizing mental health in schools.
Delivery of Evidence-Based Programs: A foundational component of the MTSS framework involves the delivery of evidence-based programs that have been shown to be effective in promoting mental health and addressing mental health problems in school settings.
Increasing Pressure on Schools: There has been increasing pressure placed on schools to import evidence-based prevention and treatment programs in response to students' mental health needs (Langley, Nadeem, Kataoka, Stein, & Jaycox, 2010), underscoring the growing recognition of the role of schools in supporting student mental health.
Growing Number of Evidence-Based Programs: There are a growing number of evidence-based programs established for use in school settings (Forman et al., 2013), providing schools with a range of options for addressing student mental health needs.
Focus of Programs: These programs typically address behavioral, social, and emotional factors assumed to cause or exacerbate disruptive, noncompliant and aggressive behavior (Wilson & Lipsey, 2007), although a growing number address mental health more broadly (e.g., emotion regulation, trauma, depression, anxiety), reflecting the expanding scope of school-based mental health interventions.
Modalities of Programs: Feature a variety of modalities including classroom-wide support systems and behavioral health curricula, small group socio-emotional skills training and peer support, and comprehensive, multicomponent programs that typically integrate training for child, parent, and teacher (August, Bloomquist, Realmuto, & Hektner, 2007; August, Realmuto, Winters, & Hektner, 2001), offering a diverse range of approaches to address student mental health needs.
Standardized Delivery: To standardize and facilitate delivery, these programs are generally delivered with uniform composition, dosage, and duration to students regardless of their individual risks and needs (August, Gewirtz, & Realmuto, 2010), raising concerns about the appropriateness and effectiveness of a one-size-fits-all approach.
Limitations of "One Size Fits All" Approach: This approach assumes that all children have similar needs, which may not be the case in diverse school populations with varying mental health challenges.
Modest Effect Sizes: Despite their intuitive appeal and evidence base, such programs have yielded only modest effect sizes with considerable variability in individual response (Rones & Hoagwood, 2000), suggesting the need for more tailored and adaptive interventions.
Call for More Adaptive Approaches: Such performance has led some researchers to call for more adaptive, customized approaches that are tailored to the individual needs of youth (Collins, Murphy, & Bierman, 2004), leading to the development of innovative strategies such as Adaptive Treatment Strategies (ATS).
Tailored Problem-Solving Approach: Adopting a more tailored problem-solving approach to service delivery is consistent with the basic tenets of MTSS as a proactive and responsive framework, yet efficiency and feasibility are also very real and important concerns that need to be addressed when implementing tailored interventions.
Avoiding "Program for Every Problem" Phenomenon: We must also avoid the “program for every problem” phenomenon (Domitrovich et al., 2010), which can lead to a fragmented and inefficient system of care.
Challenges in Delivering Tailored Approach: Determining how to deliver a tailored, problem-solving approach while maintaining efficiency and feasibility is a challenge that requires careful planning and resource allocation.
Guidelines Needed: While this framework offers a promising approach for providing students with the services they need, there are few guidelines for:
(a) selecting the most appropriate interventions for each tier, which requires a thorough understanding of the evidence base and the specific needs of the student population
(b) determining how best to sequence the interventions in a tiered approach, which involves considering the potential synergistic effects of different interventions and the optimal timing for intervention delivery
(c) how to determine the best intervention sequence for any individual student, which necessitates a comprehensive assessment of the student's strengths, needs, and preferences.
Differing Theoretical Orientations: These challenges are compounded by the emergence of programs developed with differing theoretical orientations, which can make it difficult to integrate different interventions into a cohesive and effective system of care.
For example, interventions implemented within the context of PBIS are grounded in behavioral principles, while interventions implemented within the context of social-emotional learning (SEL) are grounded in the principles of positive youth development, requiring a careful consideration of the underlying theoretical assumptions when selecting and implementing interventions.
Difficulty for School Professionals: These challenges make it incredibly difficult for school professionals to determine which programs to implement in their settings, and which programs will yield the greatest effects for their student population, leading to uncertainty and potential for ineffective or even harmful interventions.
Emerging Innovation: Adaptive Treatment Strategies (ATS): The present article describes an emerging innovation in the development and validation of precision-based interventions for youth who experience social, emotional, and behavioral impairments and need additional support, offering a promising approach to addressing the limitations of traditional one-size-fits-all interventions.
ATS (also known as dynamic treatment regimes) apply principles similar to those used in MTSS to tailor each individual's intervention over time based on assessment of ongoing response but extend these models in several ways, providing a more flexible and data-driven approach to intervention delivery.
ATS Specifics:
ATS specify (a) which intervention option to offer first, based on the student's individual characteristics and needs
(b) at what time point response should be assessed and interventions adjusted, allowing for timely and data-driven modifications to the intervention plan
(c) which intervention option should be offered if there is nonresponse to the first intervention option, ensuring that students receive the most appropriate and effective intervention based on their ongoing response.
Intervention options may vary in intensities, types, and/or modalities, allowing for a highly individualized approach to intervention delivery.
Sequential Multiple Assignment Randomized Trials (SMART): The construction of these decision rules is aided by an innovative research methodology called sequential multiple assignment randomized trials (SMART), which allows researchers to identify the most effective intervention sequences for different student profiles.
SMART empirically evaluates multiple intervention sequences and associated decision rules within a single trial in order to identify optimal ATS, providing a rigorous and data-driven approach to developing adaptive interventions.
Article Overview: In the text that follows we:
(a) present the rationale for ATS, highlighting the limitations of traditional interventions and the need for more adaptive approaches
(b) describe the SMART technology used to operationalize ATS, providing a detailed explanation of the methodology and its application in intervention research
(c) describe a SMART prototype currently being delivered by a community agency to preempt the development of conduct disorder among at risk youth, illustrating the practical application of SMART in a real-world setting
(d) describe an example of how schools might apply a SMART to evaluate multi-tiered interventions to prevent or deescalate behavior problems, demonstrating the potential of SMART to improve the effectiveness of MTSS in schools.
Adaptive Treatment Strategies (ATS)
ATS Defined: ATS use ongoing information about an individual (e.g., changes in behavioral status) to make subsequent intervention decisions through the use of decision rules (i.e., algorithms), ensuring that interventions are tailored to the individual's evolving needs and progress.
Decision Rules: Decision rules specify how the composition and/or intensity of an intervention should be adjusted at critical decision points such as when an individual is not responding to a current intervention (August, Piehler, & Bloomquist, 2014; Lei et al., 2012), promoting a flexible and data-driven approach to intervention delivery.
Resemblance to Real-World Practice: ATS resemble ‘real-world’ practice where practitioners often change treatments when an individual fails to demonstrate a desired response, absent empirically established decision rules, but ATS provide a more systematic and evidence-based approach to treatment modification.
Recommendations Based on Individual Characteristics: With ATS, recommendations for adjusting treatment are based on individual characteristics that are collected and measured during treatment, such as “has the individual exhibited significant symptom reduction” or “has the individual reached a specified level of adaptive functioning?”, ensuring that intervention decisions are informed by objective data and tailored to the individual's specific needs.
Readjusting Intervention Plan: When an individual displays no response or possibly a suboptimal response, practitioners may readjust the intervention plan by increasing dosage or switching to a different intervention, promoting a flexible and responsive approach to intervention delivery.
Time-Varying Approach: This time-varying approach is particularly useful for the treatment of chronic disorders such as depression (Lavori, Dawson, & Rush, 2000), Attention-Deficit Hyperactivity Disorder (ADHD; Pelham et al., 2016), and alcohol and drug dependence (Krantzler & McKay, 2012), where individuals may require ongoing adjustments to their treatment plan over time.
Prevention Tool: ATS may also have a role as a prevention tool by redirecting risk trajectories for children or youth deemed to be at heightened risk for conduct disorder (August et al., 2014), offering a proactive approach to preventing the development of serious mental health problems.
Natural Fit with MTSS: Systematically incorporating empirically-derived ATS within the MTSS framework appears to be a natural fit, as both approaches share a common goal of tailoring interventions
With both approaches, the goal is to adapt intervention options based on an individual's progress to a specific goal, promoting a personalized and data-driven approach to service delivery.
Four-Stage Problem-Solving Model: Underlying the MTSS framework, an individually-oriented four-stage problem-solving model is frequently utilized (Kratochwill & Bergan, 1990), providing a structured approach to addressing student needs.
These stages entail identification of a problem, problem analysis, intervention implementation, and intervention evaluation, ensuring that interventions are carefully selected, implemented, and evaluated to maximize their effectiveness.
Careful Selection of Intervention Programming: This problem-solving process allows for careful selection of appropriate intervention programming based on student needs and evaluation of program effects, promoting a data-driven and evidence-based approach to service delivery.
ATS as a Systematic Extension: Indeed, in many ways ATS could be considered a systematic extension of this approach within MTSS, providing a more rigorous and data-driven framework for tailoring interventions to individual student needs.
Value Added to Incorporating ATS Paradigm: We propose that there is value added to incorporating the ATS paradigm as a mechanism for studying MTSS, as it can provide valuable insights into the effectiveness of different intervention sequences and the optimal timing for intervention delivery.
Consistency with MTSS Rationale: ATS is consistent with the underlying rationale for MTSS, with a central aim being to ensure that all youth are able to access the services they need, promoting equity and access to effective interventions.
Evaluating Interventions in Isolation: In particular, within the context of MTSS, researchers often evaluate evidence-based interventions in isolation as standalone programs, with little attention to how a series of interventions perform either within tiers or in combination across tiers of support (i.e., synergistic effects), limiting our understanding of the complex interplay between different interventions.
Disconnect Between Research and Delivery: Thus, a disconnect exists between how these interventions are evaluated and how they are delivered within MTSS, hindering the translation of research findings into practice.
Need to Bridge Research-to-Practice Gap: Such research is imperative in order to help bridge the research-to-practice gap and is consistent with growing interest in implementation science in school settings (Owens et al., 2014), which seeks to promote the adoption and implementation of evidence-based practices in real-world settings.
Theoretically-Informed Intervention Planning: Furthermore, the MTSS emphasis on theoretically-informed intervention planning is highly germane to the development and implementation of ATS, as it ensures that interventions are grounded in sound theoretical principles and address the underlying causes of the problem.
Problem Analysis Stage: Within MTSS, a problem-solving approach is often implemented, including a problem analysis stage, which involves the identification of specific, potentially malleable, causal factors maintaining a problem (Kratochwill & Bergan, 1990), providing a foundation for selecting targeted and effective interventions.
Informing Intervention Selection: In turn, this analysis informs the selection of more appropriate and precise interventions, ensuring that interventions are tailored to the individual's specific needs and the underlying causes of the problem.
ATS and Intervention Programs: Similar to intervention selection with the problem-solving approach, intervention programs utilized within ATS typically consist of interventions that target common theoretically-derived mechanisms of problems, promoting a targeted and theoretically-sound approach to intervention delivery.
Empirically-Derived Decision Rules: Within ATS however, empirically-derived decision rules utilizing individual characteristics may guide the selection of intervention programming, providing a data-driven and personalized approach to intervention selection.
Example Assessment of Problem Behavior: For example, an assessment of problem behavior may reveal specific established etiological pathways (e.g., social skill deficits or poor emotion regulation) contributing to such behaviors in an individual student, providing valuable information for selecting targeted interventions.
Decision Rule Guided Selection: A decision rule will guide the selection of an ATS that employs an intervention empirically demonstrated to most effectively target the identified key contributing factors, ensuring that interventions are aligned with the individual's specific needs and the evidence base.
Theoretically-Derived Decision Making: Thus, ATS allow for a process of theoretically-derived decision making that is guided by empirical evidence, promoting a rigorous and data-driven approach to intervention delivery.
Student Response to Intervention: Finally, a hallmark of MTSS and the associated problem-solving approach involves student response to intervention, yet very little is known regarding non-responders in the current landscape, highlighting the need for more research on effective interventions for students who do not respond to initial interventions.
Research on Non-Responders: Research examining more precise and appropriate interventions for students who are unresponsive to intervention is critical to advance prevention efforts, as it can help to identify and address the factors that contribute to non-response and develop more effective interventions for these students.
Pioneering Work in Reading: A number of researchers have pioneered work in this vein, particularly in the area of reading (e.g., Connor et al., 2009; Torgesen, 2000), providing valuable insights into the factors that contribute to reading difficulties and the development of effective interventions.
Less Attention to Social, Emotional, and Behavioral Domains: However, in educational settings, social, emotional and behavioral domains have received considerably less attention, highlighting the need for more research on effective interventions for students with social, emotional, and behavioral challenges.
Evidence-Based Practices for Non-Responsive Students: It could be argued that, in many ways, evidence-based practices do not currently exist for non-responsive students, as traditional research methodologies often focus on identifying interventions that are effective for the average student, rather than on identifying interventions that are effective for students who do not respond to initial interventions.
Little Attention to Non-Responders in Traditional Research: Indeed, within traditional group-based research methodology, little attention is typically paid to non-responders, limiting our understanding of the factors that contribute to non-response and the development of effective interventions for these students.
Positive Main Effect Concerns: That is, when an intervention is found to produce a positive main effect, there are likely students participating in the intervention group who either (a) did not change or (b) actually declined in performance, highlighting the importance of attending to the needs of non-responders.
Concern for These Students: We should attend to and be concerned about these students and seek to develop more adaptive, precise and effective interventions, ensuring that all students receive the support they need to be successful.
Precision-Based Care: Such precision-based care is at the conceptual core of an MTSS framework: many consider MTSS to be a needs-driven, equity based approach to ensuring that all children receive the supports they need to be successful, promoting equitable access to effective interventions.
Trial-and-Error Approach: Instead, in practice, a trial-and-error approach to modifying subsequent efforts is typically employed, which can be inefficient and may not result in the most effective intervention for the student.
Elements School-Based Teams Change: For example, school-based teams my decide to change any number of elements, including:
(a) the format of delivery for students unresponsive to intervention (small group versus individual), tailoring the intervention to the student's learning style and needs
(b) dosage frequency or intensity, adjusting the intervention to the student's level of need and response
(c) implement a different intervention entirely, ensuring that the student receives the most appropriate and effective intervention.
Promising Approach: Adaptive treatment strategies offer a promising and rigorous approach to the study of intervention delivery and tailoring to promote positive outcomes for all students, but particularly for those with the greatest need, providing a framework for developing and implementing effective interventions for all students, regardless of their level of need or response to initial interventions.
SMART Technology
ATS Implementation in Schools: While schools implementing MTSS might employ ATS in their delivery, the embedded ATS may not have undergone rigorous evaluation and optimization, limiting their effectiveness and potential for improving student outcomes.
Questions to Address in Constructing ATS: In constructing ATS, questions that need to be addressed include the best sequencing of interventions when individuals are not responding and the best time to evaluate response, ensuring that the ATS is tailored to the individual's needs and the specific context.
SMART Design: Construction of such high-quality ATS can be achieved with an innovative type of research design referred to as sequential multiple assignment randomized trials (SMART; Almirall et al., 2014; Lavori & Dawson, 2008; Murphy, Oslin, Rush, & Zhu, 2007), providing a rigorous and data-driven approach to developing and evaluating adaptive interventions.
Multiple Stages and Randomizations: A SMART design is implemented in multiple stages with individuals randomized multiple times to various intervention options across stages (see Lei et al., 2012), allowing researchers to compare the effectiveness of different intervention sequences and identify the optimal intervention for each individual.
Balanced Groups: Sequenced randomizations ensure that at each decision point, the groups of participants assigned to each of the intervention options are balanced in terms of participant characteristics, reducing the risk of confounding variables and ensuring that the results are valid.
Randomization Example: In the examples provided below each youth is randomized twice, initially and then again once it is known whether the youth is a responder or a non-responder to the initial intervention, allowing researchers to identify the optimal intervention for each individual based on their response to the initial intervention.
SMART Experimental Design & Randomization in ATS: It is important to note that even though the SMART experimental design involves randomization, once ATS have been developed, their delivery in customary practice does not involve randomization, as the decision rules are used to guide intervention selection based on the individual's characteristics and response.
Key Time Points: Each randomization stage within the SMART becomes in the ATS a key time point where a decision is made as to whether to adjust the intervention, allowing for timely and data-driven modifications to the intervention plan.
Empirically-Derived Decision Rules: Data from the SMART are used to construct empirically-derived decision rules that may be utilized within these time points as a part of the ATS, ensuring that intervention decisions are based on objective data and tailored to the individual's specific needs.
Selection of Appropriate Interventions: The selection of appropriate interventions is critical in the creation of effective ATS, as the interventions must be aligned with the individual's needs and the underlying causes of the problem.
Possible Intervention Options: Possible intervention options may reflect different types of conceptual orientations (behavioral-based contingency systems vs. socio-emotional skills training), different intervention foci (youth vs parent), different modes of delivery (classroom, small group, individual, parent/family), different levels of dosage/intensity, and/or different approaches to increase engagement and adherence to the intervention, providing a wide range of options for tailoring interventions to individual needs.
Operationalizing Decision Rules: The goal is to operationalize decision rules such as—begin with intervention type X, if the individual shows favorable response, step-down to maintenance; if the individual shows a poor response, step-up to intervention Y, providing a clear and actionable framework for intervention decision-making.
Multiple ATS Derived from One SMART: Depending on the number of decision points and the number of intervention options to consider at each decision point, multiple ATS can be derived from any one SMART, allowing researchers to identify the optimal intervention sequence for different student profiles.
Answering Key Tactical Questions: Set up in this way, a SMART allows researchers to answer key tactical questions, such as “What is the best first stage intervention option,” “What second stage intervention option is best for individuals who do not show satisfactory response to the first stage intervention option,” “Which sequence of intervention options yields the best outcomes?”, providing valuable insights into the effectiveness of different intervention strategies.
Examination of "Downstream" or Synergistic Effects: This approach allows for an examination of “downstream” or synergistic effects whereby an initial intervention component enables an individual to benefit more substantially from subsequent intervention components (Murphy et al., 2007), promoting a more comprehensive understanding of the complex interplay between different interventions.
Systematically Identifying Effective Approaches: By evaluating different first-stage, second-stage, and overall sequences of interventions, a SMART allows us to systematically identify which approach may be most effective in addressing the diverse needs and risk factors of a population, ensuring that interventions are tailored to the specific needs of the population.
Tailoring Variables: In addition to tailoring intervention sequences based on assessment of ongoing response, SMART can further increase precision by identifying potential tailoring variables that reflect pre-intervention individual characteristics, allowing for a more personalized approach to intervention delivery.
Measuring Personal Characteristics: By measuring personal characteristics before initiating services, researchers may evaluate these characteristics for their utility in predicting differential response to the various intervention sequences, providing valuable information for matching individuals with the most effective interventions.
Secondary Tailoring Variables: If successfully identified, these personal characteristics could then serve as secondary tailoring variables to match individuals with their optimal intervention, further enhancing the precision and effectiveness of the intervention.
Potential Secondary Tailoring Variables: Potential secondary tailoring variables may include gender, age, SES, as well as biomarkers, personality traits and psychosocial risk factors, providing a wide range of options for tailoring interventions to individual needs.
Integrating Secondary Tailoring Variables: When validated in SMART, these secondary tailoring variables can be integrated into the derived ATS, further enhancing the precision and effectiveness of the intervention.
Answering the Question: “What Works Best for Whom?”: As such, these data help answer the question, “What works best for whom?”, ensuring that individuals receive the most appropriate and effective interventions based on their unique characteristics and needs.
Application of ATS Within and Across MTSS Tiers: ATS derived through SMART may be applied both within and across MTSS tiers, providing a flexible and adaptable framework for intervention decision-making.
Within Tiers: Within tiers, an ATS may inform the section of an optimal intervention strategy for a particular youth at that level of support. Strategies within tiers may be selected based on knowledge of a best intervention option within a combination of interventions offered. Individual tailoring variables may also be used to select an optimal intervention strategy for a particular youth within a tier of support.
Across Tiers: Across tiers, ATS will provide a standardized approach to evaluate non-response within a particular tier and the appropriate subsequent more intensive intervention to be offered at a higher tier.
Consistent Empirical Guidance: By providing decisions rules dictating intervention selection, identification of non-responders, and intervention sequence, ATS have the potential to build more consistent empirical guidance across the MTSS framework.
Executing a SMART Design
Juvenile Diversion Agency: The Community Prototype
Description: This section presents a description of a community-based implementation of a SMART design (see August et al., 2014).
Intent: The intent is to provide the reader with the rationale and approach for constructing ATS as a programming framework for a high risk youth population (diversion youth) and an illustration of how a SMART design was crafted to operationalize ATS.
Collaborative Partnership: A collaborative partnership between a youth-serving agency and a university-based team of research investigators implemented this project, leveraging the expertise of both practitioners and researchers to develop and implement effective interventions.
Agency Role: The agency serves the county attorney's office by providing pre-court diversion programming for juvenile offenders who have been cited by law enforcement for various status and misdemeanor offenses including shoplifting, vandalism, disorderly conduct, underage drug use, and assault but have not yet established a pattern of serious and chronic antisocial behavior or have been formally adjudicated, providing an opportunity to intervene early and prevent the escalation of delinquent behavior.
Diversion as a Portal: Diversion serves as a key portal for identifying at-risk youth, many of whom are at heightened risk for developing Conduct Disorder (CD) as well as depression, anxiety disorders, and substance use disorders (Wareham, Dembo, Poythress, Childs, & Schmeidler, 2009), allowing for early identification and intervention to prevent the development of serious mental health problems.
Standard Practices for Diversion Agencies: The standard practices for diversion agencies are to require restitution, community service, or simply to warn-and-release, which may not be sufficient to address the underlying causes of delinquent behavior.
Limitations of Punitive Interventions: Although these punitive interventions satisfy the public demand for accountability, there is little evidence that they prevent escalation of conduct problems or progression to more serious mental disorders (Patrick & Marsh, 2005; Wilson & Hoge, 2013), highlighting the need for more effective and evidence-based interventions.
Alternative: Life-Skills Approach: An alternative to restorative justice programs is a life-skills approach that emphasizes the acquisition of strengths and the building of human capital in order to maximize the likelihood that youth offenders will veer away from a delinquent lifestyle toward more conventional goals (Guerra, Williams, Tolan, & Modeski, 2008), providing youth with the skills and resources they need to succeed in life.
Emphasis on Evidence-Based Programs: In light of the increasing emphasis on evidence-based programs for prevention and treatment of youth conduct problems, the collaborative community agency/research team opted to select interventions with an evidence-base for reducing conduct problems, ensuring that the interventions are effective and aligned with best practices.
Effective Interventions: There is overwhelming evidence that youth problem-solving skills training and/or parent behavioral management skills training are effective interventions in this domain (Kazdin, 2010), providing a foundation for selecting targeted and effective interventions.
Challenges Faced by Diversion Counselors: The collaborative further recognized that diversion counselors confront additional intervention challenges as a result of the heterogeneity in the diversion population, requiring a flexible and adaptable approach to intervention delivery.
Heterogeneity in Diversion Population: The diversion population includes youth who may be one-time offenders and at minimal risk for future offending, periodic offenders who are at moderate risk for future offending, as well as chronic offenders who are at heightened risk for serious and chronic offending, requiring a range of interventions to address the diverse needs of the population.
Limitations of "One Size Fits All" Approach: Consequently, conventional interventions that apply a “one size fits all approach” are unlikely to be effective in addressing these diverse individuals, highlighting the need for more tailored and adaptive interventions.
Negative Effects of Incongruent Interventions: Moreover, providing interventions to youth that are incongruent with their needs and motivations can result in low levels of engagement and dropout (Kazdin & Wassell, 1999) as well as negative peer contagion effects particularly when programs are provided using a group format (Dodge, Dishion, & Lansford, 2006), underscoring the importance of tailoring interventions to individual needs and preferences.
Need for New and Creative Approaches: It stands to reason that new and creative approaches are needed to effectively address the diverse needs of the diversion population.
Guidelines for Effective and Efficient Intervention Framework: With these considerations in mind, the collaborative produced the following guidelines to assist in their efforts to produce an effective and efficient intervention framework:
Youth (and their families) referred by law enforcement for juvenile diversion programming are not help-seekers in the traditional sense. Thus, youth and their families must perceive intervention options offered to them as relevant to their needs and administered with minimal burden, ensuring that they are engaged in the intervention process.
In light of the varying degrees of risk for continued offending among a diversion population (heterogeneity), an adaptive intervention approach in which intervention options are tailored to the youths' risk profiles may produce the best outcomes, providing a personalized and data-driven approach to intervention delivery.
An efficient and cost-effective adaptive intervention approach would feature a sequential, stepped care delivery system in which youth receive only the intervention(s) they need, minimizing the burden on youth and families and