JMIR Publications

ABSTRACT

Background:

Tailoring intervention content, such as those designed to improve physical activity (PA) behaviour, has been shown to induce and enhance effects. Previous research has shown that it might be relevant to tailor PA interventions based on gender; results revealed important differences between gender-based subpopulations with respect to roles of determinants in the intervention working mechanism. In order to optimise tailoring, one needs to understand the differences between subpopulations based on other characteristics.

Objective:

In this study, by means of Bayesian networks, differences in PA intervention working mechanisms of subpopulations based on the relevant moderators age, education level and presence of physical impairments are investigated.

Methods:

This study analyses subpopulation-specific subsets of an integrated dataset from five PA intervention studies, including demographic factors, an indication of control versus experimental group, and baseline, short- and long-term measurements of PA and its socio-cognitive determinants. The relevant subpopulations are defined based on the factors age, education level and presence of physical impairments. For each subpopulation, a stable Bayesian network is estimated based on the corresponding subset of data by applying a bootstrap procedure, and, according to a confidence threshold, relevant paths of the model are visualised in order to find indications regarding subpopulation-specific intervention mechanism.

Results:

Comparison of subpopulation-specific models unveils similarities and differences with respect to determinants’ roles in PA behaviour change induced by interventions. Similar structures of determinants affect short-term PA, ultimately causing effects in the long-term, where intention and habit are directly related to PA for most subpopulations. With respect to age-based differences, the interventions influence PA less via attitude cons and planning for older than younger people. Looking at level of education, less influence arises through planning and intrinsic motivation for low educated compared to high or medium educated participants, whereas more influence takes place through attitude pros for this low educated group with respect to maintaining effects in the long-term. Looking at physical impairments, apart from the findings that attitude pros and planning are more relevant in the pathway of change for people without impairment, an even more interesting insight is that substantial fewer determinants are directly influenced by the intervention within the group with impairments.

Conclusions:

Intervention mechanisms in specific demographic groups are rarely studied so far. (Our initial) interpretations from the derived subpopulation models in this study unveil subpopulation-specific patterns of behavioural change, that enable us to better tailor intervention content to characteristics of the target population in order to induce or enhance effects.