Novel Statistical Models for Intensive Longitudinal Analyses of Cancer Control Behaviors

点击翻译按钮获取中文摘要

PROJECT SUMMARY This proposal responds to PAR-18-352 (“Methodology and Measurement in the Behavioral and Social Sciences”). It will develop, apply, test, and disseminate multilevel statistical models and software for estimating effects of intraindividual means, variances, slopes comprised of time-varying variables on cancer control risk factors and outcomes. Cancer remains a leading cause of mortality. Approximately 42% of new cancer cases in the U.S. are viewed as potentially avoidable including 19% caused by smoking and 18% caused by excess body weight, physical inactivity, excess alcohol consumption, and poor nutrition. Intensive Longitudinal Data (ILD) methods, which collect many assessments captured at high density on a micro-timescale (e.g., seconds, minutes, hours) using real-time data capture methodologies (e.g., mobile sensing and accelerometry), offer enormous opportunities for insight into the dynamic nature of cancer risk factors and outcomes. In ILD studies, it is common to have hundreds to thousands of observations per subject, and this allows us to model intraindividual parameters comprised of time-varying variables such as means (e.g., how unhappy is a subject, on average, across occasions?), variances (e.g., how erratic is a subject’s mood across occasions?), and slopes (e.g., is a subject’s mood related to feelings of energy across occasions?). In our preliminary work, we have begun developing a series of two-stage multilevel statistical models to test intraindividual mean, variance, and slope parameters as predictors, mediators, and moderators of subject-level outcomes. We have shown that greater intraindividual variability in, but not mean of feeling energetic, was associated with lower odds of meeting physical activity guidelines. We have also found that intraindividual variability in positive affect moderated the associations of intraindividual mean positive affect with depressive symptoms and alcohol consumption. However, current data analysis techniques are limited in several key areas, which severely limit our ability to capitalize on the full potential ILD to enhance behavioral and social science research in cancer control. To address these gaps, we propose the following aims. Specifically, we will develop multilevel models capable of (1) estimating effects of intraindividual variances and slopes for time-varying variables nested in time or within clusters of people, (2) predicting cancer control outcomes nested within time or clusters, (3) comparing the relative predictive strength of several intraindividual variances, and (4) examining intraindividual variances or slopes for count or ordinal variables. We will test these statistical extensions by conducting secondary analyses of data from three cancer control studies using ILD (with 94,892 occasions nested within 643 people) including children, middle-aged adults, parent-child dyads. We will also develop, test, and disseminate a stand-alone software with GUI capable of running these statistical models to be used by applied behavioral and social science researchers. The methods to be developed can easily generalize to a variety of other disease areas such asthma, disordered eating, suicide prevention, HIV risk, medication adherence, and environmental exposures.