Can behavior change techniques be delivered via short text messages?

Abstract

Despite significant advancements in behavioral science it is unclear whether behavior change techniques (or BCTs) can be delivered to large numbers of people in a cost-effective and reliable way. The current study investigated whether it is possible to reliably deliver BCTs using short text messages. Short text messages were designed to deliver each of the 93 BCTs specified in the BCT taxonomy v1. Following initial coding and refinement by the team, a Delphi study with a panel of 15 experts coded which BCT each short text message was designed to deliver and also rated whether they were likely to be understood by recipients and easily converted to target different behaviors. After two iterations, the experts correctly assigned 66 of the 93 messages to the BCT that they were designed to deliver and indicated that these messages were likely to be easy to apply to a range of behaviors and understood by recipients. Experts were not able to identify which BCT 27 of the messages were designed to deliver and it was notable that some clusters of BCTs (e.g., "Goals and planning") were easier to deliver via short text messages than other clusters (e.g., "Scheduled consequences"). The findings suggest that short text messages can be a reliable way to deliver many, but not all, BCTs. The implications of the current study are discussed with respect to the delivery of specific BCTs and clusters of the taxonomy, as well as the need to test the acceptability of interventions delivered via short messages and the impact of messages on behavior.