JMIR Publications

ABSTRACT

Background:

Several recent trials have examined automated short message systems (SMS, i.e., ‘text messaging’) to provide remote breastfeeding support to mothers, but vary in terms of their design features and outcomes examined.

Objective:

In this paper, we examine user engagement with and feedback on a theory-grounded SMS intervention intended to prevent perceived insufficient milk (PIM)—the single, leading modifiable cause of unintended breastfeeding reduction and cessation.

Methods:

We recruited 250 nulliparous women intending to breastfeed between 13-25 weeks of pregnancy in southwestern Pennsylvania. Women were randomly assigned with equal allocation to either a SMS intervention to prevent PIM and unintended breastfeeding reduction or cessation (A mobile, semi-automated text message-based intervention to prevent perceived low or insufficient milk supply, i.e., MILK; n=126) or a control group receiving general perinatal SMS-based support via the national, free Text4Baby system (n=124). Women in both groups received text messages 3-7 times per week from 25 weeks of pregnancy to 8 weeks postpartum. The MILK intervention incorporated several automated interactivity and personalization features (e.g., keyword texting for more detailed information on topics, branched response logic), as well as an option to receive one-on-one assistance from an on-call study lactation consultant. We examined participant interactions with the MILK system, including response rates to SMS queries. We also sought participant feedback on MILK content, delivery preferences, and overall satisfaction with the system via interviews and a remote survey at eight weeks postpartum.

Results:

Participants randomized to MILK (70% white, 68% college-educated) reported that MILK texts increased their breastfeeding confidence and helped them persevere through breastfeeding problems. Nine women (7%) elected to stop MILK messages and three (2%) opted to reduce message frequency during the course of the study. There were 46 texts through the MILK system for individualized assistance from the study lactation consultant (54% on weekends or after-hours). The most commonly texted keywords for more detailed information occurred in postpartum weeks 4-6 and addressed milk volume/intake and breastfeeding/sleep patterns. MILK participants stated a preference for anticipatory guidance on potential breastfeeding issues and less content addressing benefits of breastfeeding. Suggested improvements included extending messaging past eight weeks, providing access to messaging for partners, and tailoring content based on women’s pre-existing breastfeeding knowledge and unique breastfeeding trajectory.

Conclusions:

Prenatal and postpartum evidence-based breastfeeding support delivered via semi-automated SMS is a feasible and acceptable intervention for first-time mothers. To optimize engagement with digital breastfeeding interventions, enhanced customization features should be considered. Clinical Trial: NCT02724969, “A Mobile, Semi-automated Text Message-based Intervention to Prevent Perceived Low or Insufficient Milk Supply (MILK)”