INTERVENTION DESIGNS

What is an adaptive intervention?

Adaptive interventions (AIs) are multi-stage treatment pathways that guide clinicians and educators on how best to respond to the changing needs of individuals.

The result is better long-term outcomes for the greatest number of people.

Why adaptive interventions?

Adaptive interventions address the challenges of heterogeneity.

Not everyone benefits equally from the same treatment. Not everyone is receptive to treatment at the same time or in the same environment. Adaptive interventions take human differences into account both as treatment begins and during the course of treatment. This ensures each person receives treatment that continuously works for them.

Adaptive interventions respond to practical constraints.

Not all treatments are available at all times. Some treatments are more resource-intensive than others. Adaptive interventions allow practitioners to maximize the efficiency of treatment by delivering the most resource-intensive interventions only when people will benefit from them.

When clinicians and educators deliver interventions that recognize the diversity of human behavior and biology, health outcomes improve across all populations.

What are the components of an AI?

Not everyone benefits equally from the same treatment. Not everyone is receptive to treatment at the same time or in the same environment. Adaptive interventions take human differences into account both as treatment begins and during the course of treatment. This ensures each person receives treatment that continuously works for them.

What does an adaptive intervention look like?

A schematic.

Adaptive interventions are frequently represented as schematics, with later-stage intervention options branching out from earlier-stage options along a timeline. Labels with advancing arrows indicate what conditions activate which intervention options. Labels on the timeline may indicate the activation of decision points based on criteria other than time.

A series of decision rules.

Adaptive interventions are built on conditional premises—like algorithms. They are sometimes also represented as a series of if-then statements known as decision rules.

Same shape, many sizes.

An adaptive intervention takes place over two or more stages. The developer decides the duration and number of stages.

What does an adaptive intervention look like?

A schematic.

Adaptive interventions are frequently represented as schematics, with later-stage intervention options branching out from earlier-stage options along a timeline. Labels with advancing arrows indicate what conditions activate which intervention options. Labels on the timeline may indicate the activation of decision points based on criteria other than time.

A series of decision rules.

Adaptive interventions are built on conditional premises—like algorithms. They are sometimes also represented as a series of if-then statements known as decision rules.

Same shape, many sizes.

An adaptive intervention takes place over two or more stages. The developer decides the duration and number of stages.

What makes adaptive interventions unique?

AIs take in multiple types of information.

Traditional treatment models tailor interventions based on a narrow band of static information—age, race, gender, socio-economic status. Adaptive interventions capture a broader spectrum of both dynamic and static information to account for how people change during multiple stages of treatment.

What makes adaptive interventions unique?

AIs take in multiple types of information.

Traditional treatment models tailor interventions based on a narrow band of static information—age, race, gender, socio-economic status. Adaptive interventions capture a broader spectrum of both dynamic and static information to account for how people change during multiple stages of treatment.

AIs respond to multiple types of information.

Decision rules within an AI may call upon static information, dynamic information, or some combination of the two to determine which intervention option should be delivered at a given stage.

Where are adaptive interventions found?

Adaptive interventions play a critical role in many fields. These include precision healthcare, where there is a need to link interventions to concrete determinants of health, and education, where academic frameworks call for intensive, evidence-based support for students. In education, adaptive interventions are sometimes part of multi-tiered systems of support (MTSS) or positive behavioral interventions and support systems (PIBS) used in schools. In medicine, adaptive interventions are often used as part of “clinical decision support” systems. Wherever they are used, adaptive interventions go by many different names.

Constructing adaptive interventions.

The d3center specializes in developing new methods for optimizing (constructing) adaptive interventions. In this context, optimization refers to the construction of an effective and efficient adaptive intervention. The methods developed by d3c also include new experimental approaches for the design and analysis of sequential, multiple-assignment, randomized trials (SMARTs).

Adaptive Intervention Resources

References

Almirall, D., Nahum-Shani, I., Sherwood, N. E., & Murphy, S. A. (2014). Introduction to SMART designs for the development of adaptive interventions: with application to weight loss research. Translational Behavioral Medicine, 4(3), 260-274.

Collins, L. M., Murphy, S. A., & Bierman, K. L. (2004). A conceptual framework for adaptive preventive interventions. Prevention Science, 5(3), 185-196.

Nahum-Shani, I., & Almirall, D. (2019). An Introduction to Adaptive Interventions and SMART Designs in Education. NCSER 2020-001. National Center for Special Education Research.

Nahum-Shani, I. & Militello, L.K. (2018). Promoting Military Family Well-Being with Digitally Supported Adaptive and Just-In-Time Adaptive Interventions: Opportunities and Challenges. The National Academies of Sciences, Engineering, and Medicine consensus study: The Well-Being of Military Families; Commissioned paper: https://www.nationalacademies.org/our-work/the-well-being-of-military-families

Pfammatter, A. F., Nahum-Shani, I., DeZelar, M., Scanlan, L., McFadden, H. G., Siddique, J., … & Spring, B. (2019). SMART: study protocol for a sequential multiple assignment randomized controlled trial to optimize weight loss management. Contemporary Clinical Trials, 82, 36-45.

LET’S STAY IN TOUCH

Join the d3center Mailing List

Keep up to date with the latest news, events, software releases, learning modules, and resources from the d3center.