Welcome back to the Coursera course, how to apply the multiphase optimization strategy, MOST, in your intervention development research. This is the beginning of Module 3, Introduction to the optimization trial. This video is Lesson 1, titled Importance of the resource management principle. I'm Linda Collins of the School of Global Public Health, New York University. I will be your narrator for this module and I am also one of the course developers. The other developer is Kate Guastaferro of the College of Health and Human Development, Penn State. This is a figure you've seen before. It provides an overview of MOST. As you've learned in Modules 1 and 2, MOST has three phases, namely preparation, optimization, and evaluation. In this module, we will be focusing on the optimization phase. We will emphasize one type of optimization trial design, the factorial experiment. Later in this module, we will briefly review a number of optimization trial designs. In this lesson, you will learn how to recognize the importance of basing the selection of an optimization trial design on the resource management principle. But first, why conduct an optimization trial at all? The objective of an optimization trial is efficient assessment of intervention components-- that is, how each component performs on average and how it performs in combination with other components. You might consider an optimization trial if you want to: weed out underperforming intervention components, get a sense of the magnitude of each component's affect, or examine whether the effect of a component is augmented or reduced in the presence of another. Of course, statistically, this is an interaction. This information is then used to select components or component levels to optimize the intervention and thereby achieve intervention EASE. We discussed intervention EASE in Module 1. Intervention EASE is a strategic balance of the four desiderata for multicomponent interventions. In other words, a strategic balance of effectiveness against affordability, scalability, and efficiency. The choice of design for the optimization trial is critical. Any experimental design is a possibility, but it must be selected based on the resource management principle. Let's review the resource management principle. In Module 1, you learned that according to this principle, an investigator using MOST must strive to make the best and most efficient use of available resources when obtaining scientific information. Any experimental design is okay as long as it is the most efficient one. You may be thinking, how do I recognize the most efficient experimental design? The most efficient design for an optimization trial is the one that: best manages research resources (we will be discussing this soon), AND most directly addresses the primary research questions. The types of primary research questions may depend partly on the type of intervention to be optimized. We're going to discuss this later in this module. Let's refresh our memory about the hypothetical example we've been using. In this hypothetical example, an intervention scientist is developing an intervention aimed at reducing viral load among HIV positive heavy drinkers. Candidate components are: motivational interviewing, no or yes, peer mentoring, it could be no or yes, text message support, no or yes, mindfulness meditation, again, it could be levels no or yes, and behavioral skills training, which is a little bit different. The levels here are low intensity and high intensity. We're going to use this as an example throughout this module but please note that sometimes we may use a subset of the candidate components to illustrate different points. Here's some of the logic underlying the resource management principle. In theory, a huge, maybe if you had five components, 32-arm RCT would be definitive about which combination of components is best, but it's not feasible to power. Instead, we'll manage research resources strategically so as to gain the most information, gain the most reliable information, and move intervention science forward the fastest. This means that the intervention scientist has to decide what information is most important to obtain and target resources there. It's the responsibility of the intervention scientist to choose the design most appropriate for the type of intervention being optimized-- again, we're going to discuss this later in this module-- and choose the design for efficiency. Note that the starting point is the resources you have or can reasonably apply for. By definition, MOST does not require an increase in research resources, but in most cases, it will require a realignment of research resources. In other words, you may find you're budgeting differently and spending money on different things. Let's discuss the groundwork an investigator must lay before selecting an experimental design. The objective here is to gather information that will be used in decision-making, that is, in making decisions about which components and component levels go into the optimized intervention. Essentially, you need as much practical information as possible. The starting point is always, what decisions do I need to make? Let's return to our example for a moment, and consider what decisions have to be made for each of these five components. Let's review the decisions that need to be made for candidate components, 1-4. For each of these components, the question is, does the component have a detectable effect? That is, does the yes level outperform the no level? If an effect is detected, the component will be considered for inclusion in the intervention package. In other words, a difference between yes and no levels has been detected, and therefore, the yes level will be selected. I said the component will be considered for inclusion because depending on the optimization objective, which was discussed in the previous two modules, one or more effective components may have to be left out to make the intervention affordable or scalable. I'd like to comment that this is a very simplified discussion of decision-making. For example, we have not specified what kind of effect we're talking about, and we're not considering interactions. But for now, let's focus on the big picture. If no effect is detected, the component is not included in the intervention. In other words, the no level is selected. The behavioral skills component, as I said before, is a bit different because its levels are not no and yes. Instead, they're low-intensity and high-intensity. Here the question is, does the high-intensity level outperform the low-intensity level? If high-intensity does outperform low intensity, the high-intensity level would be under consideration for inclusion in the intervention. Depending on the optimization objective, this may be reconsidered in relation to cost broadly defined. In other words, it is possible that to make the intervention affordable, it might be necessary to settle for the low-intensity level. If high-intensity is not better than low-intensity, the low-intensity level will be selected. Here's what we want to find out by conducting the optimization trial. For each component, we want to determine whether there is a difference between the higher and lower levels. We did not discuss this, but ideally, we would want to determine whether components interact. This information is used in making decisions about selection of components and levels for the intervention package. The resource management principle says that the investigator must carefully choose an experimental design so as to gather the information needed while making the best use of, but not exceeding, the available resources. Thus, the experimenter must have a clearly-specified set of research questions, know what resources are available to conduct the research, and know what resources are required by each experimental design under consideration, because different experimental designs can require different resources. We're going to demonstrate this in a later lesson in this module. The resource management principle suggests that to select an experimental design, it's a good idea to consider several alternatives and examine the scientific information each will provide, and whether that information is what you want. Also, what each design costs in terms of number of experimental subjects and number of experimental conditions. The starting point is the resources you have or can reasonably apply for, and the idea is to gather the greatest quality and quantity of information you can get with those resources. In this course, we're emphasizing the factorial optimization trial. Why? Well, it's the most widely applicable optimization trial, and it's often the most efficient. Moreover, most, but not all, other optimization trial designs are rooted in the factorial experiment, so it's a good idea to be familiar with and understand it. In this lesson, you learned how to recognize the importance of basing the selection of an optimization trial design on the resource management principle. In the next lesson, you will learn how to describe the factorial experiment, and what it is intended to estimate, that is main effects and interactions, and to recognize that coding can have implications for interpretation of results. See you then.
