JMIR Publications

ABSTRACT

Background:

For the next generation of interventions to exploit eHealth’s full potential, they should build on theories of eHealth’s distinct working mechanisms. Interventions have traditionally built on models and theories from other fields: static behavior change theories or face-to-face therapy. However, eHealth interventions are neither static nor do they rely on human contact. Nevertheless, few efforts have thus far been made to identify the distinct working mechanisms of eHealth. We suggest that a promising and pragmatic research method for this purpose is the qualitative interview. However, getting interesting interview data on eHealth working mechanisms can be surprisingly challenging, posing a threat to study quality. Furthermore, there are to date no guidelines for enhancing the quality of such studies.

Objective:

This paper’s aim is to (1) describe five inherent challenges to interview studies on eHealth working mechanisms, (2) suggest tools to counteract each challenge, and (3) suggest an explanation for these challenges that also may increase our understanding of eHealth interventions’ unique features.

Methods:

We started with the problems encountered in one specific interview study, upon which began an analytic process of looking for underlying causes, abstraction and generalizing. Through comparisons with other studies and consulting the qualitative literature, we agreed upon a refined set of challenges and tools.

Results:

We describe five challenges: achieving a joint understanding, not straying off the interview topic, recalling program experiences, seeing through the social interview situation, and mixing applied and basic research. Then we present tools for clarifying and exhausting the research topic, keeping contextual answers short, aiding recall, arranging and analyzing the interview situation, and structuring the dual-aim interview.

Conclusions:

We suggest that there may be an underlying cause of these challenges. Presumably, change processes are influenced by the program both through its content and through its interaction with the user (e.g. tailoring, tunneling, providing reminders). However, the program is not seen as an actor, and therefore the interaction is largely invisible. This invisible interaction makes it difficult for the participant to see how his/her change processes have been affected by the program, cloaking its working mechanisms. The suggested interview tools serve to render the invisible person-program interaction visible again, so that the interaction and its influence on the participants’ change processes can be explored. The tools offered in this paper may be a step towards more fully developed guidelines to enhance the quality of future studies – which in turn can provide a foundation for field-specific theorizing that can enable program developers to harness eHealth’s distinct working mechanisms in the future.