BACKGROUND
Interventions delivering information through the internet and via text messages to mobile phones have been widely adopted as a means to reach parents with health and parenting information and support. However, retaining participants, once enrolled, remains a challenge. The SMS4dads text messaging program, developed in Australia, has delivered texts to over 10,000 fathers. While acceptance of the program is good, almost one in five ask to be removed. Analysing and reporting attrition from digital parenting programs such as SMS4dads may assist in developing more effective interventions.
OBJECTIVE
Examine multiple attrition measures in a text-based intervention targeting fathers.
METHODS
Demographic characteristics, program features, participant views, and para data were examined to identify attrition factors among fathers.
RESULTS
Participants who were smokers, recorded risky alcohol consumption, had a lower education level, or signed up antenatally, had 30%-47% higher hazard of dropout from the program while participant age, Aboriginal and/or Torres Strait Islander status, rurality, and psychological distress score (K10 category) were not associated with dropout. Of those who replied with a reason for dropping out (202/605; 33.4%) almost half cited ‘Other reasons’ (n=83, 41.1%), followed by ‘Not helpful’ (n=47; 23.3%) and ‘Too busy’ (n=44; 21.8%). Fifty two percent of inactive participants (those who had not clicked on any links or Mood Checker for at least 10 weeks) confirmed that they wanted to remain in the program, 18% chose to drop out, and 29% remained in the program.
CONCLUSIONS
Analysing program features can provide additional guidance for improving program adherence. In this case, repeated surveys and specific message content were not related to drop out while demographic features such as low education were. As well, seeking feedback from inactive participants allowed an estimate of dropouts among this group. Planning for evaluations of attrition from health and parenting text-based programs may need to be done early in the design phase of the research.