|
Home  Research Areas Multiphase Optimization Strategy
Behavioral interventions for prevention and treatment are an important part of the fight against drug abuse and related conditions such as HIV/AIDS and Hepatitis C. The science of behavioral interventions in these areas faces many challenges. Among them is how to optimize the potency of interventions in order to achieve the greatest public health benefits.
Generally, even the best behavioral interventions have not been systematically optimized. This suggests they could be performing even better. Optimization of new interventions as they are being developed, and revision of existing interventions with the objective of optimization, holds out the promise of substantial improvements in public health impact, and a corresponding reduction in morbidity and mortality.
Our perspective is that it is important both to optimize and evaluate behavioral interventions. Evaluation of behavioral interventions, which is much more familiar to most investigators, is typically conducted using the standard two-arm treatment/control randomized controlled trial (RCT). Our work suggests that optimization of behavioral interventions requires a different approach. This is why we have developed a methodological framework called the Multiphase Optimization STrategy (hereafter called MOST; Collins, Murphy, Nair, & Strecher, 2005; Collins, Murphy, & Strecher, 2007). MOST is based on optimization principles originally developed in engineering and routinely used there and in other fields.
It is important to note that MOST is a framework for conducting intervention optimization research rather than a procedure that can be taken off the shelf and applied as-is in any given situation. Some key features of MOST are:
- MOST is based primarily on randomized experimentation. As much as possible, randomized experiments are used to gather information, and the results of these experiments form the basis for decision making.
- MOST is a phased approach. Scientific information is gathered in systematic steps. In any particular step, the information gathered is used to make the decisions needed to proceed to the next step. This means that sometimes an experiment will be followed by another.
- Experimental designs are chosen carefully using a resource management perspective (Collins, Dziak, & Li, in press). This means that designs are chosen so as to maximize scientific information gained while minimizing resource expenditures. Although any design that meets these criteria may be used, factorial designs tend to be very economical and thus are frequently the design of choice.
- All decisions that are made by the investigator, including experimental design choices, are informed by the behavioral theory underlying the intervention. Thus this theory must be clearly articulated before MOST begins.
References:
Collins, L.M., Murphy, S.A., Nair, V., & Strecher, V. (2005). A strategy for optimizing and evaluating behavioral interventions. Annals of Behavioral Medicine, 30, 65-73.
Collins, L.M., Murphy, S.A., & Strecher, V. (2007). The Multiphase Optimization Strategy (MOST) and the Sequential Multiple Assignment Randomized Trial (SMART): New methods for more potent e-health interventions. American Journal of Preventive Medicine, 32, S112-S118.
Resources:
Contacts:
|
