Innovative methods for observing and changing complex health behaviors: four propositions

Abstract

Precision health initiatives aim to progressively move from traditional, group-level approaches to health diagnostics and treatments toward ones that are individualized, contextualized, and timely. This article aims to provide an overview of key methods and approaches that can help facilitate this transition in the health behavior change domain. This article is a narrative review of the methods used to observe and change complex health behaviors. On the basis of the available literature, we argue that health behavior change researchers should progressively transition from (i) low- to high-resolution behavioral assessments, (ii) group-only to group- and individual-level statistical inference, (iii) narrative theoretical models to dynamic computational models, and (iv) static to adaptive and continuous tuning interventions. Rather than providing an exhaustive and technical presentation of each method and approach, this article articulates why and how researchers interested in health behavior change can apply these innovative methods. Practical examples contributing to these efforts are presented. If successfully adopted and implemented, the four propositions in this article have the potential to greatly improve our public health and behavior change practices in the near future.

Keywords: Adaptive interventions; Computational models; Ecological momentary assessment; Idiographic; Precision health.

Figure 1

Simulated percentages of a person’s level of motivation (y-axis) modeled over time (x-axis); (A) measurement of motivation across three time points, representing conventional intervention evaluation at baseline, post-intervention, and at a longer-term follow-up; (B) measurement of motivation on different days compared with (A) but maintaining the same measurement frequency; (C) measurements at a higher sampling frequency (40 time points instead of three); (D) linear regression line (dashed) and LOESS regression line (solid), fitted to the measurements in (C); (E) measurements at a higher sampling frequency (400 time points instead of 40), revealing a process of “deterministic chaos.” Figure courtesy of Matti Heino, University of Helsinki, see [27].

Figure 2

Simulated forest plot of within-person associations between physical activity and stress levels the following day, inspired from the results obtained in [35]. Effect sizes with 95% confidence intervals for each participant are presented on the x-axis. The dashed vertical line represents the average effect across participants. The forest plot of within-person associations allows the visualization of heterogeneity for a particular average effect. Here, physical activity is significantly associated with higher levels of stress for the first 3 participants (1–3); the association is nonsignificant for the next five participants (4–8); and physical activity is significantly associated with lower levels of stress for the remaining seven participants (9–15).