Screening Experiments and the Use of Fractional Factorial Designs in Behavioral Intervention Research

Abstract

Health-behavior intervention studies have focused primarily on comparing a new program over the control using randomized controlled trials. However, we are seeing a dramatic increase in the number of possible components (factors) due to developments in science and technology (internet, web-based surveys, and so on). These changes dictate the need for alternative methods that can screen a large set of potentially important components in order to identify the important ones quickly and economically. We have developed and implemented a multiphase experimentation strategy for accomplishing this goal. This article describes the screening phase of this strategy and the use of fractional factorial designs for studying several factors (also called components) economically and uses two on-going projects on behavioral intervention to illustrate their usefulness. The fractional factorial designs are supplemented with follow up experiments in the refining phase, so any critical assumptions about interactions can be verified and the dosage levels can be optimized.

The Effectiveness of Web-Based Tailored Smoking Cessation Interventions on the Quitting Process (Project Quit): Secondary Analysis of a Randomized Controlled Trial (Preprint)

“…Third, we present a novel analysis from a precision medicine perspective using the notion of TR; this type of analysis will help understand what type of smokers will benefit from a tailored smoking cessation intervention in terms of optimizing their quitting process. As mentioned previously, the results suggest that smokers with low education are more likely to benefit from tailored interventions; this is consistent with the prior findings based on the seven-day point prevalence data [8,[19][20]. Thus, our current analysis re-confirms the role of number of quit attempts as an additional outcome over the commonly-used point prevalence outcome.…”

“…The primary aim of the study was to identify and test the effects of five psychosocial and communication intervention components influencing smoking cessation [8]. In order to screen multiple components, the study employed a multiphase optimization strategy (MOST) [21] framework, implemented using a 16-cell (2 ) fractional factorial design [19,22] in which each of the five components (outcome expectations, efficacy expectations, success stories, message source, and message exposure) were varied at two levels, high vs. low. The study protocol was reviewed and approved by the institutional review board of each collaborating institution and of the University of Michigan in January 2004.…”

Worldwide application of prevention science in adolescent health

“…92,94 Innovative and cost-effective methods for designing and assessing programme adaptations are emerging in prevention science to guide this process. 95 …”

Multilevel Interventions: Study Design and Analysis Issues

“…J Natl Cancer Inst Monogr 2012;44: [49][50][51][52][53][54][55] intervention (individuals, neighborhood, and community) to increase influenza vaccination areas in New York City. They also did not attempt to assess the relative or incremental contribution of the different components but only the overall impact on individuals.…”

“…One then could evaluate the feasibility and cost of different designs that allow testing of those effects (41,53). Some interactions might not be worth testing because the components involved are clearly dependent in one another.…”