Modeling intensive longitudinal data with mixtures of nonparametric trajectories and time-varying effects.

Abstract

Behavioral scientists increasingly collect intensive longitudinal data (ILD), in which phenomena are measured at high frequency and in real time. In many such studies, it is of interest to describe the pattern of change over time in important variables as well as the changing nature of the relationship between variables. Individuals' trajectories on variables of interest may be far from linear, and the predictive relationship between variables of interest and related covariates may also change over time in a nonlinear way. Time-varying effect models (TVEMs; see Tan, Shiyko, Li, Li, & Dierker, 2012) address these needs by allowing regression coefficients to be smooth, nonlinear functions of time rather than constants. However, it is possible that not only observed covariates but also unknown, latent variables may be related to the outcome. That is, regression coefficients may change over time and also vary for different kinds of individuals. Therefore, we describe a finite mixture version of TVEM for situations in which the population is heterogeneous and in which a single trajectory would conceal important, inter-individual differences. This extended approach, MixTVEM, combines finite mixture modeling with non- or semi-parametric regression modeling, in order to describe a complex pattern of change over time for distinct latent classes of individuals. The usefulness of the method is demonstrated in an empirical example from a smoking cessation study. We provide a versatile SAS macro and R function for fitting MixTVEMs.

Heterogeneity in perinatal depression: how far have we come? A systematic review

“…The intensive observations of an individual symptom over time offer unique opportunities to describe detailed temporal changes and identify related environmental and psychosocial antecedents and consequences at a finer level, and recent advancements in statistical modeling (Dziak et al , 2015) also provide suitable tools to uncover fine-grained information contained in such intensive data. However, it is also highly desired to simultaneously model multiple relevant symptoms using the intensively collected data to gain knowledge on the etiology paths of depressive symptoms, in addition to temporal changes of individual symptoms.…”

“…However, it is also highly desired to simultaneously model multiple relevant symptoms using the intensively collected data to gain knowledge on the etiology paths of depressive symptoms, in addition to temporal changes of individual symptoms. This would require a more advanced statistical approach that is capable of handling intensive multivariate longitudinal observations, with the aim of untangling the complex, possibly evolving, inter-correlation structure among these symptoms; the work of Dziak et al (2015) makes a good start in this direction. In addition, the emerging network analysis approach (Borsboom and Cramer, 2013), which has recently been employed to study intrinsic structure of mood disorders, might provide a promising complementary strategy to Dziak et al (2015) to tackle the ambitious task of uncovering the complex dynamic inter-correlation structure among symptoms related to PND heterogeneity.…”

“…This would require a more advanced statistical approach that is capable of handling intensive multivariate longitudinal observations, with the aim of untangling the complex, possibly evolving, inter-correlation structure among these symptoms; the work of Dziak et al (2015) makes a good start in this direction. In addition, the emerging network analysis approach (Borsboom and Cramer, 2013), which has recently been employed to study intrinsic structure of mood disorders, might provide a promising complementary strategy to Dziak et al (2015) to tackle the ambitious task of uncovering the complex dynamic inter-correlation structure among symptoms related to PND heterogeneity. Another possibility is to call for network analysis on the group level and try to derive predictors for positive vs negative distal outcomes (Boschloo et al , 2016, van Borkulo et al , 2015).…”

Negative affect subtypes and craving differentially predict long-term cessation success among smokers achieving initial abstinence

“…These results provide new evidence for the motivational significance of negative affect in relapse to smoking and the validity of negative affect reduction as an important target for smoking cessation interventions. Finally, it should be noted that it is generally assumed that one of the effects of NA is to increase craving (Dziak et al 2015). Thus, statistically controlling for craving when examining NA-withdrawal association may have resulted in an underestimation of important variance accounted for by affective symptoms.…”

Extreme Response Style and the Measurement of Intra-Individual Variability in Affect

“…Intra-individual variability in affect can be studied by attending to the within-person variance in PA and NA sum scores across visits. There is growing interest in intra-individual variability in affective symptoms amongst smokers making quit attempts (e.g., Geiser, Griffin & Shiffman, 2016) arising from findings that such variability is substantial (Dziak et al, 2015; McCarthy et al, 2006), and that it may significantly influence the success of quit attempts (Piasecki et al, 2003a, b). Thus, in addition to the effects of mean affect, greater intra-individual variability in affect that emerges in the post-quit period is anticipated to increase the likelihood of relapse to smoking (see e.g., Hedeker, Mermelstein, Berbaum, & Campbell, 2009; McCarthy et al, 2006; Piasecki et al, 2003a, b).…”