The aims and themes of this proposal reflect mechanisms of single and multiple behavior change with an emphasis on cancer risk behaviors to add to our understanding of effective behavior change interventions that can promote behavior change and public health while reducing healthcare costs. The efficacy of tailored programs currently available is still not strong enough, and one of the most important barriers to enhancing the efficiency and effectiveness of behavior interventions is insufficient knowledge about the mechanisms of behavior change. Such mechanisms have been viewed, essentially, as a black box with unknown contents. Conducting empirical research using a variety of new methods to target this 'black box' and elucidate its contents is critical for progressing to the net generation of intervention systems. The proposed study will employ latent class analysis (LCA) and latent transition analysis (LTA) to model the complex mechanisms of single and multiple behavior change in four cancer risk behaviors, including smoking, unhealthy diet, sun exposure, and inactivity. Secondary data analyses will integrate data from four population-based intervention trials targeting (a) two different samples of parents of adolescents (N's = 2,460, and 2,547), (b) one sample from patients from an insurance provider list (N = 5,382), and (c) worksite employees (N = 1,906). All four randomized trials targeted smoking, diet and sun exposure, and one trial of parents and one of employees targeted exercise in addition to the other three behaviors. All trials used comparable TTM-tailored interventions and the full assessment to participants who were at risk for the behaviors at baseline. The participants were assessed also at 12 and 24 months follow up. A series of LCA and LTA analyses will sequentially address the specific aims of this proposal: (1) to examine progression through the stages of change for individual cancer- related risk behaviors; (2) to examine progression through the stages of change for multiple (i.e., pairs or three) behaviors; (3) to examine stages of change algorithms for multiple (pairs or three) behaviors, as well as the potential predictors for stage membership using multivariate statistical modeling techniques; (4) to compare the four risk behaviors in terms of the stage progressions; (5) to compare the applicability of different complementary analytical methods, including LCA and LTA, and manifest variable approaches in terms of understanding mechanisms of behavior change. The potential insights from this study may provide an empirical foundation for development of more effective, low cost, tailored interventions for multiple risk behaviors and determine the potential LCA/LTA has as an alternative analytical framework in examining behavior change, as well as advancing the science of cancer prevention.
This project will examine mechanisms of single and multiple behavior change in four cancer risk behaviors, including smoking, unhealthy diet, sun exposure, and inactivity. This will utilize latent class analysis (LCA) and latent transition analysis (LTA) examining pooled existing datasets from four studies on more than 12,000 participants. The potential insights from this study may help to develop even more effective, low cost, tailored interventions for multiple risk behaviors and to demonstrate the potential LCA/LTA has as an alternative analytical framework for examining behavior change and advancing the science of cancer prevention.
|Yin, Hui-Qing; Rossi, Joseph S; Redding, Colleen A et al. (2014) Validity and stability of the decisional balance for sun protection inventory. J Skin Cancer 2014:190541|