Text Messaging Interventions

See also Behavior Change Taxonomy

References

• Onighi, F. (2020). The effectiveness of text messaging interventions for improving addiction outcomes. Private Manuscript. View here.

• Rothman, A. J., Desmarais, K.-J., & Lenne, R. L. (2020). Chapter Two - Moving from research on message framing to principles of message matching: The use of gain- and loss-framed messages to promote healthy behavior (A. J. B. T.-A. in M. S. Elliot (Ed.); Vol. 7, pp. 43–73). Elsevier. https://doi.org/https://doi.org/10.1016/bs.adms.2019.03.001
• Rojewski, A. M., Duncan, L. R., Carroll, A. J., Brown, A., Latimer-Cheung, A., Celestino, P., Sheffer, C., Hyland, A., & Toll, B. A. (2020). Quit4hlth: a preliminary investigation of tobacco treatment with gain-framed and loss-framed text messages for quitline callers. Journal of Smoking Cessation, 15(3), 143–148. https://doi.org/DOI: 10.1017/jsc.2020.17

Notes re Rojewski by @Florin:

From the study Discussion section (GF = gain-framing, LF = loss framing):

• More participants in the GF group reported making a 24 h quit attempt compared to the LF group.

• The primary analysis on abstinence rates revealed no differences between groups, but the responder analysis showed a between-group difference of 9% benefitting the GF group. Although this preliminary study was not powered to identify between-group differences, a point prevalence difference of this size is a clinically significant effect on smoking cessation in the context of a quitline.

• However, not all text messages were tailored to the participant’s demographics, and the messages had broad content categories. Thus, a participant’s ‘reasons for quitting’ may not have been matched with the topic areas of the text messages, and the messages may not have been perceived as relevant to their motivation to quit. For instance, if the gain-framed message described saving money and the participant was more focused on improved health, then these messages may have had less of an impact. Indeed, GF participants reported that their messages were not as tailored as the LF participants, and the messages may therefore have had less of an effect.

• Quitline callers who receive GF messages are more likely to make a quit attempt, but they are not more likely to remain abstinent long term. This could be due to several features of the messages: (1) the messages do not line up with a participant’s self selected quit day because the participant selected a quit date further than 2 weeks out from the initial quitline call, (2) the participant has attempted to quit but has relapsed and the messages do not offer relapse support, or (3) if a participant successfully achieved abstinence, the messages did not specifically address barriers to remaining abstinent. Future research could tailor the messages to current smoking status and quit date changes.

Overall,

• even if the intervention had only achieved more quitting attempts I believe that is a great outcome given this is such a simple intervention
• as the authors note, there can be many reasons for not remaining abstinent and messages could have been better tailored
• Brickman2017 also found that gain frames were preffered over loss frames https://www.tandfonline.com/doi/abs/10.1080/14681811.2017.1332582
• This is a new science but framing and tailoring/matching messages are great areas to explore further, and in general, interventions that require no/minimal effort from the side of the client (Candeo, Hands Off etc. all used some level of tailoring).

See also

• Behavior Change Taxonomy