Table 3 Influence of formative findings on intervention design

Message timing and content:

∙ The FGD findings supported initiating messaging during pregnancy rather than waiting until postpartum.

  ∘ Weekly automated messages started at enrollment in the trial, which occurred during the third trimester of pregnancy (at/after 28 weeks).

  ∘ Automated messages stop at study exit at 6 months postpartum.

∙ Some SMS were shortened in response to feedback from male participants.

∙ Despite divergent views from men on anticipatory guidance about contraceptive physiologic or side effects, especially irregular bleeding, these messages were included.

  ∘ Perspectives on such anticipatory guidance will be explored further in post-RCT in-depth interviews with male and female intervention participants.

∙ We developed tailored messaging tracks for individual women who were single/did not desire partner involvement and for couples.

Male engagement:

∙ While the FGD findings support the acceptability of including male partners, male partners must not be recruited without the express informed consent of the woman.

∙ Women who desired to consult with male partners prior to referring them for trial participation were given this option.

∙ Men received detailed information about the intervention and also signed informed consent prior to receiving SMS.

SMS platform adaptation:

∙ In response to the pervasive side effect concerns and misperceptions, framed with narratives of contraceptive discontinuation, the SMS platform was adapted to allow for method-specific messaging “tracks” after method choice.

  ∘ E.g.: After a woman/couple choose the injection, subsequent messaging track is designed to support continuation through engagement around effects, adherence, and option to change methods.

  ∘ Method-specific tracks also eliminate non-personalized messages about other methods once a method has been initiated.

∙ In consideration of promoting couple communication, a feature was added to the system to allow for forwarding of participant messages to his/her partner if appropriate.

  ∘ E.g.: If a male partner sends in an SMS to the nurse requesting more information about the method, this SMS could be forwarded to the female partner, allowing both members to interact.

  ∘ This feature is not automatic; the clinician can decide whether it is appropriate to forward specific messages.