There is ML (Meal Data) standard domain introduced in SDTM IG 3.3.
However, a question is who is a consumer of breastfeeding and who is a study subject? If subjects of the study are babies, then you should utilize ML domain. If subjects of the study are mothers, then it would be some Findings or Events domain depends on Topic of the collected data.
it sounds to me like you might need 2 records in ML - to cover feeding with BREAST MILK and FORMULA ?
if so, I would approach as follows:
MLTRT = BREAST MILK
MLPRESP = Y
MLOCCUR = Y/N = response to Q1
MLENDTC = response to Q2
MLTRT = FORMULA
MLPRESP = Y
MLOCCUR = Y/N = response to Q3, although response is inverted. i.e. exclusively breastfed = Y implies that Formula = N
MLSTDTC = response to Q4, i.e. formula starts on the date that exclusively breast milk stops
I’ve talked to other people of our team. It was an interesting comment, that babies breastfeeding data should be mapped into EX domain in Nutrition studies.